MonkeyBrains seeks $350,000,000 for its own satellite

The happy mutants at MonkeyBrains, the San Francisco hacker-friendly ISP, have launched a $350,000,000 IndieGoGo campaign to buy their own satellite ("North Korea just launched a satellite; we want to as well"). Some fun facts about MonkeyBrains: it was founded by Rudy Rucker, Jr (son of the archduke of mutantcy, cyberpunk writer Rudy Rucker [Sr]); it is the basis for the fictional ISP in my novel Little Brother; and they want $350,000,000. Also: if the satellite thing doesn't work out, they want to use the money to fill San Francisco with high-speed fiber optics that aren't run by crappy telcos.

A quick internet search reveals that this is the cost for getting a satellite into orbit:

* Satellite manufacture: $150M
* Satellite launch: $120M
* Launch insurance: $20M
* In-orbit insurance: $20M
* Satellite operations (15 years): $15M

Faster Internet!

Our initial research seems to indicate having a satellite in orbit may not speed up your internet at all. []. However, if more research doesn't bode well for a geostationary satellite, we will take all of the $325M to fund either:

* Fiber to the home.
* A balloon tethered to the Farallon islands.
* a hovering drone over the Bay.

MonkeyBrains satellite


  1. If only you could get them in the same room with Kim Dot Com. Submarine cables, satellites, orbiting giant lasers, all the base are belong to them, MUAHAHAHA!!!!

  2. Unless your intent is to provide internet-of-last-resort to the sticks, there is no way a satellite is a good idea. The latency is appalling, and it’s a laws-of-physics issue, so you are stuck with it. For any reasonably dense area, even WISPs and cellular data look like god’s gift to connectivity compared to satellite… 

    1. Back of envelope:
      Speed of light is 175,000 miles per second.
      Geo-synchronous satellites are ~25,000 miles up.

      If you had a laptop, stood next to the uplink, and it was directly under the satellite, the first up-down is 50k miles making the round trip a total of 100k.

      Today’s hardware and terrestrial-based networks are fast so lets pretend away penalties and it’s still over a half-second of latency for a requested URI no matter how wide the bandwidth. Useless for VOIP and most multiplayer games. Terrible user experience for websurfing in general.

      TL;DR: I agree with your analysis.

  3. Perhaps they should just skip the satellite idea and go straight to fiber. Otherwise they’d be no better than the crappy ISPs they want to replace.

    1. Perhaps they should just skip the satellite idea and go straight to fiber.

      But then they’d need to build a space elevator to get the fiber up there.

  4. We are researching a string of mini-satellites (à la Zac Manchester) in a low earth orbit that will pass directly over the city.  Basically, we would have a ‘string of pearls’ orbiting around the globe to provide data to SF via satellites.  This will eliminate latency issues and line-of-sight issues traditional WISPs have as all dishes will be pointed straight up (don’t forget to drill a drain hole unless you want a bird bath).

    1. If the chain of satellites is at the atmosphere’s edge (62 miles but lets bump it to 87 miles) the round trip for a connection is 350 miles, about .0002 seconds so that’s nice. I’m guessing that with the more complex flight path (compared to GEO) fuel burn will be higher so satellites will de-orbit sooner. Add in that time over target will be short as satellites will be moving fastest during that section of orbit and I would think this will require a lot of satellites (is it possible to keep some in reserve that will take the place of the de-orbiters?) to maintain uninterrupted connections.

      Good luck.

      P.S. Sorry if it seems like I’m crawling up your ass with this. However, once SlashDot gets this they’re going to chop the plan into tiny pieces and consider all the parts in minute detail.

      1. The radiation pattern of our earth station antennas will be approximately 90 degrees north-south and 1 degree east-west.  Orbit will be higher — things tend to fall back to earth at 87 miles up — closer to 10ms round trip.  Youtube/netflix/your email will be cached in orbit.

          1. Visualize: many satellites along a path beaming down at SF… many clients (all on the ground) watching that narrow path in the sky that all the satellites will track.  The ‘eye’ of the antenna will see the width of the path (plus 1 degree) and 90 degrees of the sky (when a satellite leaves that window, another one will enter on the other end).  If economical, the ground stations will have a ‘phased band’ instead of a ‘phased array’ to pick out the satellite in that 90 degree window.

          2. Getting a constellation of non-GEO, non-equatorial satellites to ascribe the same track trough the sky as viewed from a fixed point on the ground is a non-trivial problem.  (GEO and other equatorial orbits are ruled out by the 10 ms latency requirement.)  None of the existing LEO constellations (Iridium, Global Star, Orbcomm) does that.  They rely on omni or minimally directional earth station antennas.  Have you actually worked out the set of orbits (inclination, # planes, # sats) that would be needed to meet this requirement?

    2.  Ummm…LEO satellites need tracking antennas, which makes them much more expensive and failure prone, or omni antennas, which have poor performance.

      1. Our plans call for a phased array.  Using beam forming techniques (activate different antennas and/or activating multiple antennas with signals slightly out of phase) we can accurately aim our beam at San Francisco.  Also, we will be able to zap other cities in the future that fall along the satellites’ pathway.

        1. PAAs are fine for the satellite but get spendy for mass-market earth stations.  I was primarily concerned the earth stations, which you have addressed in another sub-thread.

          1.  :)  Thanks, but I have a couple of my own satellites that I need to finish which will probably keep me busy through 2016.  Always happy to talk shop with new satellite ventures though.  Good luck.

    3. We are proposing a chain of condominiums (à la Manchester) in a geostationary orbit that will sit directly between communications satellites.  Basically, we would have a ‘string of users’ orbiting around the globe to consume services over IP via satellites.  This will eliminate latency issues and line-of-sight issues traditional residents have as all dishes will be pointed straight ahead (no need to drill a drain hole as there aren’t too many birds up there).

    4. Another point…$350M is a reasonable price for a big GEO satellite like ViasSat-1 or Jupiter-1, which cover all of the densely populated areas of the US.  A more limited GEO project could be done for half that.  LEO projects, on the other hand, tend to be much more expensive.  Many more satellites are necessary and those satellites spend most of their time doing nothing – 70% of the Earth is covered in water while only a few % of the Earth is densely populated, so only a few minutes out of each orbit is spent operating at capacity.  As a result, the total amount of hardware required to provide a given amount of throughput is much higher at LEO than GEO. 

        1. Specifically, start reading at the “buy the orbital assets of the bankrupt suckers who tried it, for pennies on the dollar, then succeed!” step and go from there.

    5. A string of small satellites in low Earth obit? Your not thinking of threatening to deorbit them over major cities (San Francisco appears to be the prime target) and blackmail the world’s governments from the safety of your volcano-base?

        1. So if Boing Boing were to hatch such a plot from the safety of our lair, we could count on your support?  I’ll have one of the minions make a note of that.

  5. The first thing they’ll spend the money on is buying IndieGoGo, which is worth rather less than the 7% commission on $350M. 

    Gotta love their style, though. 

  6. We used MonkeyBrains at work.  When we switched offices, we left for Comcast and haven’t looked back. Hacker-friendly, maybe.  But the service and support was shit.  Sometimes we’d get a call back when service was restored, sometimes not.  My favorite downtime was due to someone at MB mysteriously unplugging something that apparently needed to be plugged in.  My least favorite involved someone in the background shouting “I don’t know what’s wrong with it!”

    They’re cute in the scrappy underdog sort of way, but not the sort of group I’d want to spearhead a satellite launch.  I’d much rather see them get their act together with their existing services (and expand coverage in SF).

    1. My favorite downtime was due to someone at MB mysteriously unplugging something that apparently needed to be plugged in.  My least favorite involved someone in the background shouting “I don’t know what’s wrong with it!”

      I’ve never had Internet service with MB, so I can’t really comment on how their total experience is compared to other providers. But in these two cases, the only difference between MB and any other ISP (including Comcast and AT&T) is that you knew of it, not that it happened.

    1. I’ve never used MonkeyBrains but I can guess a few things that I would want as a hacker(in the sense of making/fixing not illegal/breaking-in) and customer in general:

      – Not blocking common ports (80, 443, email, etc) so one can maintain their own website/servers is a good start.
      – Keep spying on customers to the minimum of the law and requiring full legal process before divulging that info to authorities.
      – Good download/upload ratio like 1/1 or 2/1 instead of consumption driving 10/1 or 50/1.
      – If static IPs aren’t available maximize DHCP lease times.
      – probably lots more

      There are millions of ways in the networking world for the Comcasts and Charters to fuck with you trying to squeeze more money from your already thinned wallet.

      Edit: no explanation of hacker-friendly on their site but here’s a service description

  7. A similar sum would get you a Skylon. See penultimate slide on…

    This gets you a re-useable SSTO craft with good physics. You also get a bitchin’ wikkid-hot riiide. If you fancy kickin’ serious Mekon ass, you could pimp it with some worthy laaasers, pew pew pewpewpew neeeeyouuurrr…

    Ahem. Sorry about that. Got a bit carried away there. And that sum doesn’t actually get you a whole working Skylon. But if they pass the hat around, I’m in.

Comments are closed.