My friend and I have started tinkering on our cars together.
He’s got an early 90s 525i BMW and I’ve got a 2000 SR5 Toyota 4Runner.
The engineering of my Toyota is so much simpler and easier to work on than his BMW. But his BMW has given us a lot more practice, if you know what I mean ^_^
"Big Money Waster" is what we used to call them. They are notorious for electronic faults in particular. They're just very complicated, which electric cars aren't.
React 18 was released in March 2022 though. It’s been over 2 years, not 3 weeks. I’ve also actually thought the React team is pretty careful about breaking changes. Their 19 breaking changes doc is mostly about removing already-deprecated APIs.
As much as I dislike Musk, this is really not true at all. It is a order of magnitude cheaper post-SpaceX to launch things into orbit then it was pre-SpaceX
> As much as I dislike Musk, this is really not true at all. It is a order of magnitude cheaper post-SpaceX to launch things into orbit then it was pre-SpaceX
Indeed - not only has SpaceX lowered the cost to get payload to space, they've also pushed other companies like ArianeGroup and ULA to lower their prices in order to compete.
Wouldn't the cost of anything only significantly decrease after a second competitor becomes competitive? Why would SpaceX decrease costs further when they are already the cheapest? Unless the increased market gained by going lower increases total income after expenses, it would be net negative to do that.
> Wouldn't the cost of anything only significantly decrease after a second competitor becomes competitive? Why would SpaceX decrease costs further when they are already the cheapest?
That's certainly the textbook economics answer.
And yet, SpaceX has had industry leading low prices since they debuted the F9. And that was true even when they had a de facto monopoly on medium-heavy launch when both ULA and ArianeGroup discontinued their rockets without having working replacements.
As to why - I guess you'd have to ask them.
It does seem to have helped to enlarge the market. I think SpaceX does on the order of 30 non-Starlink launches every year, and a lot of those are commercial. But I think everyone recognizes that they could probably charge more, and make more money if they wanted to.
That's a silly assertion. Fuel can only cost so much.
Starship is about as large as a 747; Falcon is quite a bit smaller. Neither is completely full of fuel, but fully fueling a 747 takes a couple hundred grand.
Even if Starship is 100% fuel, you're talking about $1M or so max. Less than even one of the engines.
> NASA pays the same price per seat in the Dragon as the Soyuz because they prefer not to fill up all the seats.
That is incorrect. There are now a maximum of 4 seats in Crew Dragon.[1]
While it's true that NASA had plans to take down 6 Astronauts in an emergency, 2 of them would have basically been strapped to cargo pallets. Not something NASA would engage in under normal circumstances.
---
1. > After SpaceX had already designed the interior layout of the Crew Dragon spacecraft, NASA decided to change the specification for the angle of the ship’s seats due to concerns about the g-forces crew members might experience during splashdown.
> The change meant SpaceX had to do away with the company’s original seven-seat design for the Crew Dragon.
> “With this change and the angle of the seats, we could not get seven anymore,” Shotwell said. “So now we only have four seats. That was kind of a big change for us.”
Paying for a full-service crew launch service including ground handling for payload and crew, space suits, life support, docking, and retrieval of the crew/capsule on landing is very different than paying for kg of payload launched to orbit.
The latter has gotten significantly cheaper.
NASA's price to SpaceX for the crew missions also includes development costs of the capsule and suits because there wasn't one on the market available for NASA to use.
And above all of that, price to a customer, especially a government customer with a lot of specific requirements and paperwork, is not the same as the actual cost.
Occasionally i've stumbled upon some neat tool or beautiful software and i'm like, wow - who's behind this? And then I realize it's these two folks. Their approach is so surprising and inspiring, thanks for putting out some cool stuff into the world!
The first paragraph gives a good overview of the idea:
> Permacomputing encourages the maximization of hardware lifespan, minimization of energy usage and focuses on the use of already available computational resources. It values maintenance and refactoring of systems to keep them efficient, instead of planned obsolescence, permacomputing practices planned longevity. It is about using computation only when it has a strengthening effect on ecosystems.
That's the idea. However, my initial criticism of the way permacomputing is formulated are:
1. We could have examined each of the 12 Permaculture Design Principle and attempted to directly apply them to software design. For example, "Observe and Interact" is so broadly useful and versatile (and the core of adversarial domains, such as warfare), it can easily be applied to software. You won't see it directly listed here: https://permacomputing.net/Principles/
2. The permaculture ethical principles are not there in full. "Care for life" refers to "Care for Earth", "Care for People", but nothing about "Fair Share". Comparing these two ways of looking at it, I don't see how the permacomputing formulation is an improvement on how the permaculture ethical principles are formulated. Furthermore, I think this has more to do with not sufficiently delving into the place of technologies within a regenerative paradigm. I am speculating here with little basis, but I don't think the people who came up with this got their hands dirty with planting, nurturing, and harvesting things.
However, reading more with 100r, CollapseOS, DuskOS, there is a lot of thought put into this even if I think there are some key things missing from my experience with permaculture.
It is why my friends and I are exploring the ideas of "permatech", what is Technology's full, integrated place within a living systems world view? We have yet to come up with anything coherent yet.
That is why I had been having trouble with it. Modern high technology is a lot of exploitation.
I once heard a historian described technology as a lever. A small effort has greater gains.
However, we were looking at it from a different angle. What if civilization are not walls and cities — the division of labor so that peasants can support the ruling class — but rather, in _design_? When I first posed that, one of my friends went right into architectural design. (Which is fine, since we explored Christopher Alexander’s work).
But an example of what I was thinking of was this discovery that one of the cave paintings was probably a hunting calendar. It allows the tribe to count the number of moons when there is sufficient deer.
That’s a kind of design — a kind of permacomputing - a kind of civilization if we were to reframe it as design.
I might be stretching it there.
I have seen effective use of technology in permaculture. Digging up swales and basins make use technology, whether it is with a shovel, or feeding pigs in a way so that they can dig for you.
Oops, I meant "permacomputing". Among my friends in a private discord group, we were generalizing that to all of tech and I forgot it originated from "permacomputing".
There is a related project from permacomputing that I'd like to highlight: CollapseOS/DuskOS which has overlapping and adjacent ends with what 100 Rabbits are trying to do with UXN. I know there are attempts to port UXN to DuskOS.
I think their philosophy / research is in understanding how to build modern software that is more resilient. So from that research/exploration standpoint it makes sense to me they would go that route instead of using older tech
The messaging could be more clear, but is this substantively different from how search queries are handled when DuckDuckGo forwards them to Bing? Bing doesn’t know who made the search but could log the query itself.
Well it's a big deal to me. OpenAI have not conducted themselves in a way that makes me want to share my search intents with them, and I don't care if they or DDG tell me that what I tell them "won't be used to train AI models".
100% agree with this, especially the part about slowing down code reviews. Don’t be condescending but give your opinion directly, asking if it makes sense. I personally like “what do you think about renaming this… was thinking because…”
Funny enough, my 2007 BMW had so many issues and cost so much to maintain i used to call it a “wallet burning machine”