Startup that is trying to build an open-source satellite network | Hacker News Reader

Startup that is trying to build an open-source satellite network

satellogic.com

60 pointsm0nastic16 comments

I used to work in aerospace, specifically satellite data processing. A couple thoughts:

- Their core team does not have anyone with space experience. Success and knowledge from other sectors (looks like they all come from the security sector) does not translate. Space is a whole different can of worms. There are totally different and more constraining hardware requirements, regulations, cost levels to name a few.

They have some advisory board guys with the right experience, but they will need to be heavily involved to be useful to a team of guys coming from a security background (probably more than just as advisers).

- The costs associated with operating a satellite network are so stratospheric (badum-bum ding) that their claim about "democratizing access to data" is dubious. There is a very good reason satellite data is so damn expensive right now: it is so damn expensive to get in the first place.

They could lower some costs here and there maybe, but it would still be in the wrong order of magnitude. They would need to sell the data to a LOT of customers to be a profitable business, if they want to keep the cost of data low (this is what I assume they mean by "democratizing access). They have provided hints of some potential applications, but have they demonstrated there is sufficient existing demand for this?

Probably not. So they are essentially trying to create a new market while footing astronomical up front costs themselves. That is not an attractive risk proposition.

I hope it works because it could be very cool (space! data!). But, I remain very skeptical.

This is a pretty cool idea but I didn't see anywhere that it would be open-source. On the contrary I expect this to be super expensive, at least in the beginning. As a side note, open sourcing all of the software would be useful but not enough to get your own satellites up.

They link to their Github page which has their software in it:

https://github.com/satellogic

I agree though that just having the software isn't of much use to anyone. I just think it's very interesting.

Also, I don't know what their business model is going to be (I assume selling access to the data), but in my head I imagined time-sharing satellite access (which is probably way too complicated for them to do).

Although, it does say it's open source at the bottom of the "technology" page [1], at the top of it, it boasts how it's using "proprietary satellite technology," and they don't seem to be very disappointed by their choice in technology.

I'm honestly confused.

[1] http://www.satellogic.com/#!technology/c1w2a

the satellite design is definitely not open sourced. The camera objective size and number of satellites is not mentioned anywhere.

I know there are regulations for flying through airspace which require me to get permission depending on the type vehicle/altitude obtained. Are there similar regulations for space? Do I have to go somewhere to get permission to orbit in a certain plane or is it everybody for themselves?

IIRC, you do have to get an allocation from the UN for an orbit. There's the Space Law which restricts just about anybody from placing a satellite in geostationary orbit (http://en.wikipedia.org/wiki/Space_law), and there was another similar law which I can't seem to find atm. Not sure if LEO is restricted. IMO this law is ridiculous, no one on Earth owns space.

The law is ridiculous, but if we learned anything from our history is that if someone can access/control it - someone can own it.

Not that ridiculous - there's a lot of important stuff up there, and it's easy for someone careless or malicious to accidentally damage it. Geostationary orbits in particular are very crowded, so it's not going to be easy to get an orbital slot, and if you don't follow fairly fairly exacting stationkeeping, reliability and end-of-life requirements your satellite will drift from its assigned location and disrupt or even damage the other satellites up there.

There's a huge difference between earth orbits and deep space. The reality is that the desirable earth orbits are actually fairly crowded, and allowing completely unregulated access could definitely endanger missions. The differential speeds between orbits are so great that a flake of paint can severely damage a satellite. If we want to have useful satellite missions in the years to come we need to regulate access to space.

You're right, there should be no law- if you want a geostationary slot and another satellite is in it, knock it down. I wonder if Comcast is hiring anti-satellite operators.

It's cool that they believe community can help getting better. https://github.com/satellogic/canopus/blob/master/src/kernel...

Does anyone have an idea of how fragile these 100lb satellites are? What is the potential negative environmentally of having thousands more of these satellites orbiting the earth?

Planet labs is doing the same thing with hundreds of 3kg satellites.

Yes having thousands of these satellites in LEO is going to make it crowded up there, especially as they break up over the years.

Is there even enough room in the marketplace for all of these upstart satellite imagers?

The price is going down by 1-2 orders of magnitude. Things are looking bleak for Astrium, Digital Globe etc. The sector has long needed an adjustment.

But whether space imaging is going to be worth anything for civil markets is questionable. If drone imaging can deliver less than $1 per acre then the satellites are going to lose. Hardly anyone is going to pay $0.01 per acre for 1m GSD when you can get 1cm GSD for $1.00.

Reply on news.ycombinator.com