Ajedi32

RFC8910, from 2020: https://www.rfc-editor.org/rfc/rfc8910.html

The calculator seems to be assuming capex costs don't increase over time with traditional data centers but that they do with orbital data centers, due to hardware failure. That does make some sense in theory; terrestrial data centers don't get completely destroyed on re-entry, though I think in practice there are other costs associated with aging terrestrial infrastructure that are being omitted here.

I guess we'll see. I think though that given we're now arguing over the particular assumptions used to estimate manufacturing cost rather than over physical limitations I've sufficiently made my point.

I suspect that in practice, the "or sell products that do" part of that has had a significant negative impact on the market for open source radio firmware. But I'm not an expert in this space. How difficult is it to flash the modem on a modern cell phone? Why has nobody built a garage door opener app for Fairphone?

I don't want people to have to know how to solder to install an open source OS on their personal computing device. That sort of thing should be normal and easy.

So yeah, I guess as long as you're okay with a law like that potentially becoming very easy to bypass then I don't have a problem with it. But I think once things start going down that path, most people aren't going to be satisfied with leaving it at that; they'll hear about a bypass and start pushing to restrict the freedoms that enabled it.

Only issue with that is that enforcing it requires you to make open source smart glasses illegal, since open source software cannot be used to coerce users the way closed-source software can.

Banning harmful behavior (like secretly recording someone) is pretty defensible and not all that controversial. Banning the mere possession of generic tools that could potentially be used to engage in harmful behavior is a lot more dubious. It requires infringing the rights of a bunch of innocent people merely because they might use that freedom to do something wrong.

Sounds really complicated, but at the end of the day as long as everyone has health insurance that's still by definition universal health insurance, right? In particular, having "a group of people that benefits from it even without paying a single euro for it" seems mandatory for it to be truly universal.

It sounds to me like you're envisioning something different from universal healthcare, but it's not clear to me exactly what.

That's what universal healthcare fundamentally is; everyone pays for it regardless of whether they use it or not, and regardless of the quality of service they receive in return.

What, did you think universal healthcare meant everything would be free? That's not possible; doctors and nurses don't work for free.

Couldn't they still do that though, just with an attribution notice in the corner?

FWIW this is exactly how it works on OSM, and probably why OP is complaining people were being pedantic by telling him he "shouldn't mark a street as not walkable because there is no official sign that prohibits walking".

"Would an ordinary person want to walk here?" is pretty subjective. "Are people legally prohibited from walking here?" is pretty objective. Generally only the latter would be recorded in OSM.

OSM originated in the UK so it uses UK norms for most things. You wouldn't need to tag a motorway as not allowing pedestrians because that's already implied by the fact that it's tagged as a motorway.

What OP was (probably) doing was tagging roads that aren't motorways and that have no sign banning pedestrians with `foot=no`, which is only supposed to be used in situations where walking is illegal or impossible. Eventually more experienced mappers noticed what OP was doing and reverted a bunch of his edits, which isn't a great feeling, particularly if you've been mapping that way for a while before anyone noticed

(If OP was using StreetComplete I'd blame StreetComplete for this more than the OP. They used to have a quest for pedestrian walkability that was poorly worded, which they've since removed because it was leading to exactly this kind of confusion.)

Nevermind, I was wrong. Looks like they added a quest for this recently, and they're removing it now because it the wording of the quest didn't really match the tags it was applying when you solved it.

I'd personally never seen that quest myself, but looking at it now I can see why they're removing it.

If you were using StreetComplete I don't think you would have run into that. StreetComplete is very beginner friendly, and I don't think you can use it to mark a road as prohibiting pedestrians (which, as you discovered, is easy to get wrong if you don't understand what you're doing). You can use StreetComplete to mark whether there's a sidewalk or not, but I don't think anyone would object to you marking roads without sidewalks as not having a sidewalk.

If you like this kind of thing, https://osmstats.neis-one.org/ has a lot more stats on edits.

> I worked with a state lottery that did similar things. Their target demographic was the “every day players”

That's wild. Wasn't the whole point of having a state lottery to put the state in a better position to be able to discourage people from becoming addicted? The very idea of a "target demographic" seems contrary to that goal; the "target" should be to have nobody buying tickets.

I don't think those numbers are optimistic, I suspect they've already designed the sat in CAD and that's what the calculation spat out; it's probably quite accurate. The physics of this sort of thing are quite well known and SpaceX already has experience with the type of satellite that will be needed. There's no need for them to rely on napkin math.

Cost is a different matter of course, but it seems incongruous to me that after being confronted with the fact that your power density assumptions were off by a factor of 2x you would continue to insist ChatGPT's weird assumptions about manufacturing costs scaling linearly with mass are accurate. As was already pointed out upthread, even if all the calculator's other assumptions are correct they can turn a profit if they get the per satellite cost (excluding the cost of the GPUs) below $2 million. I think the only big question mark here is launch cost.

FWIW, SpaceX is estimating their AI sats will have over double the W/kg of what that ChatGPT summary is estimating for Starlink. https://www.spacex.com/spacexai/starmind Cost/kg is not disclosed, but I think it would be a mistake to assume there were no similar improvements there.

So ChatGPT decided to assume a constant $800/kg to estimate cost, and you're using that as evidence that cost/W doesn't decrease with scale?

Anyway, my point is that Starlink is not primarily designed to harvest as much solar power as possible the way data center sats are. I'm sure they're not making it inefficient on purpose, but what you're suggesting is like using the cost/W of a solar powered traffic camera to estimate the cost/W of a solar farm. Yes, they both use solar power, but they have entirely different design goals.

Fair point about the failure rates. I wonder what causes that, given that there are no moving parts... if it's temperature fluctuations, maybe you could just run the GPU at 100% all the time to avoid that, given that the marginal cost of power for a satellite already in space is $0?

> If you assume Starlink satellite $/W

Why would you assume that, given that high power output is not a design goal of Starlink?

My point is that SSO isn't "coming to the rescue", it's literally the entire premise of the idea (greater availability of solar power there vs anywhere on earth). I don't buy "there's no more room" as an argument when we're talking about space, even in a relatively crowded set of orbits like SSO.

Doesn't Starlink already handle all that though, just with networking racks instead of GPU racks? What's so unique about GPUs compared to other computer equipment, besides the power budget?

Certainly launch has to advance a lot for this to make sense. SpaceX is betting big on Starship getting close to $100/kg. But does satellite tech have to advance much at all to make this work? SpaceX has already successfully deployed one megaconstellation into orbit and it's extremely profitable. Isn't this just another Starlink with bigger solar panels and radiators?

The promise is that it'll make producing goods and services more efficient, like just about every other invention in human history. As always, that does put some people out of work in the short term, but it doesn't increase unemployment over the long term, and it makes all of society richer, not just the few who invented it.

Sun-synchronous orbit gets sunlight ~100% of the time.

From TFA:

> In low Earth orbit, satellites can access near-constant sunlight, generating up to eight times more solar power than on Earth.

That's obviously a reference to sun-synchronous orbit. You don't get 8 times more solar power in a low inclination orbit.

They picked a good name for fighting that. The optics of trying to outlaw heresy probably aren't great. ;)

Is Google under any legal obligation to continue releasing software updates to their competitors in the first place? Maybe there are some contractual obligations in place (I remember back when Google first started making their own hardware, other Android OEMs got very, very nervous, and Google was bending over backwards to appease them, so it wouldn't surprise me), but aside from that I don't see why they'd be under any obligation to keep releasing code to others for free. Android isn't GPL, and even if it were I believe Google owns the copyright, right?

The only way to fully free Android from Google's hands is a hard fork. Graphene OS probably doesn't have the resources for that on its own, but all the other Android OEMs put together might. Frankly I'm kind of surprised it hasn't happened yet, but I do seem to remember Google having some anticompetitive clauses in their license terms for GMS which may be preventing this... not sure if those are still in place.

Important details further down: https://grapheneos.social/@GrapheneOS/117282129725629495

So it seems like the problem isn't that the new API is Pixel exclusive, but that the first and third quarterly release patches each year are Pixel exclusive?

Yes, because glucose wetware can be held liable for malpractice if it gives you nonsensical advise.

And/or ramp up production in other countries not dependent on the strait. One way or the other the market will eventually react to stabilize prices, it just takes a while (years) for the infrastructure necessary to do that to be built.

Unlike wars, EV adoption is a slow, steady, relatively predictable process. I don't think it'll have a very noticeable impact on gas prices the way short term crisis tend to.

The much more noticeable factor will be reducing how many people even care about gas prices in the first place.