Very cool visualization, but am I seeing correctly that Starlink only covers the southern half of Canada? I guess that would still mean a lot of potential users, but also most of the major cities, which are already wired up?

This is true, but practically no one lives in the northern half so no one cares.

There's also a second stage that will cover the poles (they're required by the us government to get Alaska iirc), so eventually it will cover all of it.

> "practically no one lives in the northern half so no one cares."

I have friends who live in Iqaluit, Nunavut. It's a nice place to live with a cool community, but everything there is very expensive, including internet service. How about $399/month for 5 Mbps service, capped at 55 GB? And that's with the resident's discount! If Starlink were available here it would be an absolute game changer.

https://www.qiniq.com/internet/

Ya, but it's like them and one other family :p

Seriously though, the population of Nunavut is roughly 40k. I guess that 90% of them are potential customers so maybe I shouldn't dismiss them too early, but you can see why I think SpaceX is right to consider them (and the other northern areas) lower priority.

I'm not completely sure but I believe that area has at least 97% of the population of Canada.

There's a slider to choose the area of the circle of each satellite, so maybe mine are just way too big.

Aren't the sats flying across space? Why do they appear stationary in this visualization?

In the animation, they are moving so slowly they appear to be stationary. Try lining one up with your cursor.

Ah, okay.

But there are stripes over specific areas of interest, like Canada and Africa. Maybe it's just coincidence that I looked at it when they were there?

The really dense strips would be coincidence (those are from recent launches, the satellites in them are slowly spreading apart until they look like all the other satellites). There is always a generally high concentration of satellites at the northern and southern end of the range as a result of orbital mechanics though.

Cool, thanks! Very neat tool.

degrees from horizon?

Imagine you're standing in a flat field, facing the satellite. Degrees from the horizon is how far you would have to bend your neck to be looking straight at the satellite.

FCC filings suggest 25 is the absolute minimum for service, the final constellation is likely to use somewhere in 40-60 as the actual limit.

I've received this question a few times now, anyone have a better name? Or maybe I should just add a ? button with a diagram?

You might link to the graphic on this page as a quick reference. Should be public domain as a government work if you want to host it.

https://spotthestation.nasa.gov/message_example.cfm

https://spotthestation.nasa.gov/images/astro_horizon.png

I'm guessing based on my own experience that it's obvious if you understand why that would matter.

So the diagram may be the best option, maybe with a Wikipedia link if there's an appropriate one?

A common term is zenith angle, although that's the complement of the angle you're describing.

thanks for the elaboration.. maybe just add this explanation as a tooltip on hover or something similar?

The planet is definitely in the way of radio transmissions between you and a satellite "below the horizon" and even if it could pass through the planet that's a long way. On the other hand a satellite in a boring orbit that's directly above you is 90° from the horizon with no obstacles and as close as it'll ever be so your reception ought to be good.

An optimist might believe Starlink will work at 10° above the horizon. A pessimist might doubt anybody will get it working below 80° (ie almost directly overhead). The reality is likely to be in between and this UI lets you see the effect.

"Optimism" and "pessimism" don't really apply here. While SpaceX hasn't publicly committed to what the "operational" value will be, it's likely they had a very good idea before they flew any hardware.

SpaceX is still developing the user terminal, there might be a bit of optimism/pessimism in the degree to which they'll be able to make the user terminal target different angles (without bloating costs), though the person you're replying to definitely exaggerated the extent.

How did you track the locations of the satellites live?

> NORAD maintains general perturbation element sets on all resident space objects. These element sets are periodically refined so as to maintain a reasonable prediction capability on all space objects. In turn, these element sets are provided to users. The purpose of this report is to provide the user with a means of propagating these element sets in time to obtain a position and velocity of the space object. [1]

For the curious, here is the wikipedia page [2] with details on the US Space Surveillance Network used to obtain this data (30+ ground-based radars and optical telescopes worldwide, plus 6 satellites in orbit).

[1] https://www.celestrak.com/NORAD/elements/supplemental/ (didn't link directly to the starlink txt file on purpose, link for reference only, please don't hug the site to death)

[2] https://en.wikipedia.org/wiki/United_States_Space_Surveillan...

Exactly this.

I grabbed a set of TLEs from celestrak (which I'm serving from my site), and then I propagate them forward with satellite-js [1] to the current date 6 times a second.

[1] https://github.com/shashwatak/satellite-js