Best Twitter Ever: Twitter in spaaaace!

[andygates]

I'm going to put my spacesuit on, next stop: Earth Orbit!!
2:11 PM May 11th from TwitterBerry

From orbit: Launch was awesome!! I am feeling great, working hard, & enjoying the magnificent views, the adventure of a lifetime has begun!
about 16 hours ago from web

(astro_mike)

Two awesomenesses come out of this:

First, the heady tang of the Future boils off it like condensate from a fuel tank.  One day, everyone will post like this on the way to our hols methane-boarding on Titan.

Second, look at the Shuttle post.  "From web".  The Shuttle has a web browser. 

The Shuttle has pr0n.

Rule 34, baby.  I'll see your robot dwarf clown d/s goat felching, and raise you "in space!".

[Zipperhead]

Weightless mega-boobed babes?

It opens up interesting possibilities

[rogerzilla]

Google "The Uranus Experiment".

Astoundingly, Wikipedia also has a whole page on...

Sex in space - Wikipedia, the free encyclopedia

[TimO]

I suspect the messages are posted by someone on the ground, and forwarded from the shuttle, either over one of their data links or via voice.  Anyone connecting the shuttles data systems to the Internet without an air gap in the way needs to be taken out and shot.

Various missions have Twitter feeds these days, there have been several from previous Shuttle flights, obviously Spirit and Opportunity have theirs, Mars Phoenix did, and I think Cassini has one.  I wondered about creating one for our FGMs on Cluster, but "Another orbit collecting magnetometer vectors" every 57 hours is probably a mite boring.

If we ever manage to get a CubeSat funded and launched, I'll do a Twitter feed for that (or whatever has replaced Twitter by that point!)

[Greenbank]

And "Best Tweet Ever" surely?

[andygates]

Anyone connecting the shuttles data systems to the Internet without an air gap in the way needs to be taken out and shot.

It would be so stupid as to not even be worth declaring as stupid.

[Greenbank]

Anyone connecting the shuttles data systems to the Internet without an air gap in the way needs to be taken out and shot.

It would be so stupid as to not even be worth declaring as stupid.

The cable would make the basis for a good space elevator though...

[andygates]

In case you're concerned that Massimino is hitting IM, Facebook and Twitter hard when he should be getting ready for the second mission spacewalk, which will see him and Michael T Good tackle the space 'scope's gyro assemblies, rest assured that his tweets are sent via email to Johnson Space Center, which then posts them for him.

Awww, drat. 

[Charlotte]

You'll be telling me that it wasn't actually Beagle doing the Tweeting, next.

[TimO]

Probably not, the whole Internet in space thing hasn't really taken off yet, the standards aren't very well defined, and the latency is a real bugger for conventional comms protocols.  You try getting TCP/IP to work over a link with a twenty minute delay on responses!

[andygates]

They're working on a protocol, though, because it'll make solar system stuff a ton easier if it just chats to the Mars relay or wherever.

[TimO]

Hence my use of the phrase "not well defined".  I remember looking at some standards ages back concerning interplanetary data networks, buried somewhere on the CCSDS website (CCSDS=Consultative Committee for Space Date Systems).

[Greenbank]

Best way is to go connectionless (UDP if you are using IP) and implement the caching/resend/resiliency in the software itself.

i.e., for a 20 minute latency:-

Give each packet a unique, incrementing, number.
Provide several (differing) checksums in each packet.
Repeat sending outstanding packets every 60 seconds (throttle based on bandwidth utilisation and previous packet loss)
Receiver acknowledgements/responses include packet number of last in sequence all received fine.

...etc...

Of course, a connection-oriented protocol that does this is good, however you've got the problem of the 60 minute connection setup time (given a 20 minute latency).

</mode=i_do_network_crap_for_a_living>

[TimO]

I was trying to think of other protocols with very high latency, and the only one I could think of was defined in RFC1149, but since that only carries IP, which has best case delivery only, it doesn't define any retry or error correction mechanisms.

I know AX25 has been used over a moon-bounce path, which was probably an interesting link since the latency would be around 2.5 seconds, but there don't seem to be many cases with higher latency.

I seem to recall that communications links to drills on oil platforms use acoustic modulation of the working fluid, which presumably is relatively slow, but again probably not on this scale.

[Aidan]

today   13 minutes ago  "From orbit: At the end of my spacewalk, I had time to just look at the Earth, the most awesome sight my eyes have seen, undescribable"

[tiermat]

I was trying to think of other protocols with very high latency, and the only one I could think of was defined in RFC1149, but since that only carries IP, which has best case delivery only, it doesn't define any retry or error correction mechanisms.

RFC2549 would probably work better, don't you think, although I suppose you would get into problems trying to get the carrier into a vacuum, and the problems of re-entry of said carriers....