Slashdot Mirror
The Interplanetary Internet
Roland Piquepaille writes "This article from SPACE.com about the Interplanetary Internet (IPN) is not science fiction. It is becoming a reality, Rich Gray reports. "The IPN would form a backbone connecting a series of hubs on or around planets, ships, and at other points in space. These hubs would provide high-capacity, high-availability Internet traffic over distances that could stretch up to hundreds of millions of miles." Gray adds that all the planets and satellites in our solar system have already Internet addresses and that NASA is already communicating with its earth-orbiting missions through its internal Deep Space Network. The rest of us will have to wait until at least 2005 when IPN-equipped satellites are launched. Check this column for more details. You also can read a previous Slashdot column on the same subject.
"
Because we have 3 probes and a Spacestation that need internet access, right?
How about we focus on getting things out into space first, then we worry about being able to get pr0n to them?
http://www.talknerdy.org
from ip.c
And the optimal packet size will probably be of the order of gigabytes!
We'll need protocols that absolutely minimize the number of rounds over all other considerations.
And I wonder if your downloads will be slightly slower or faster depending on the red shift or blue shift of the server :-)
Perhaps with IP addresses being allocated to far remote locations the government will allocate more funding to quantum research?
Traditional methods of broadcast certainly won't be very useful over distances where light will take minutes to travel. Unless we make rapid advances and make quantum teleportation of electronic data possible, the email of today may become the snailmail of future generations.
It is nice that NASA is developing a communication protocol for use in their missions. It seems to have some advantages: only low-power transmitters are needed on space crafts if you have a retransmitting satellite hub around. (This is in fact done already - in missions that use a lander, the part of the ship that stays in orbit retransmit signals of the lander). Using a standardized protocol, the hubs can be used for many missions, even simultaneously.
I wish the article on space.com had focussed on this a bit more instead of popularising it by mentioning Vint Cerf and the Internet. Everyone will realize that the infrastructure is not connected to the Internet (imagine - hacked satellite, DoS attacks). Also, the protocol, which can handle large delays, will have significant differences to regular TCP/IP.
Also, I'd like to know how to NASA will solve a few problems that jump to mind. The communication hubs are only useful if they are close to the sending spacecraft. The hubs can not be put halfway earth and a planet, because of the planets rotation around the sun. So the hub would have to circle around a planet in order for future missions to that particular planet can use it.
But then NASA faces the following problems: on planets with an atmosphere, the hubs will tend to fall to the surface, so in many cases they need fuel to keep their altitude.
Secondly, the hub will only be visible from one side of the planet at the time.
Thirdly, the hub needs power, where to get it from for an extended period of time? Or are the hubs only short lived? In that case, why bother at all?
My karma ran over your dogma
Surprised that no one seems to have mentioned this book, as part of it deals with conjecture as to how a universe-wide Internet might work.
For all the whining about latency times, I don't see anyone looking for solutions! Check this out:
Quantum Networking. By using photons with quantum entanglement, it's possible to transmit data from Earth to Mars with nearly no latency.
Of course, we need a few good engineers to make it practical...
I wonder what might come of this in respect to data havens, if you can evade the law on an oil rig with at least minor success, I wonder what you could do on a satellite or another planet. Also, who will control the the data flow? Some damn good encryption might be necessary.
-Silmarildur
Yes, games are out, the round trip time being more half an hour to Mars. On a slightly more practical note, instant messaging won't be fun either.
Nor will any protocol based on TCP, or any other handshaking or window/retransmissin error-correction scheme (unless you're willing to wait a LONG time for your data).
We'll need something with serious forward error correction. And a lot of the fundamental components of the net will need a rethink before they're usable by humans at even lunar distances.
For starters, web browsing will need a local cache - of essentially the whole internet. Think spiders and massive archives, and serious redesign on sites with dynamic content.
But the pipes are necessarily too small for optimistically broadcasting everything - even if it doesn't change milisecond-by-milisecond. Serious filtering.
And most of the data for sites is the images in the ads. That just HAS to get filtered down - as will other images. Boon or bane?
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way