Moon Colonists: Expect 3000ms ping times using Lunar Net | Hacker News Reader

Moon Colonists: Expect 3000ms ping times using Lunar Net

rtfa.net

33 pointsfarkinga27 comments

I had 3000ms latency on a 1200bps SLIP connection. Damn entitled kids these days. shakes cane

Wow. I was so expecting that to get downmodded.

Ping times will be high for some sites, but I doubt Google will be one of those. As soon as this Lunar Net goes up Google will have a server on the moon faster than you can say 'light-speed delay', so all your favourite Google services and YouTube videos will be available locally with nice low ping times. It's the rest of the internet you'll have to wait for.

Yes, Google will have servers on the moon, but I doubt they'd have a mirror of all youtube.

Latency wouldn't necessarily affect watching a youtube video -- sure it would start slower, but latency != bandwidth. If you set your TCP window size high enough and could ensure low packet loss you could get some pretty fast transfers going.

I suppose that also assumes gigabit+ radio links to span the distance.

Agreed, but to mirror all of youtube is redundant, the vast majority of people only watch the most popular videos. By providing a selection of 'this weeks best' you will surely provide low latency for the majority of files. Similarly ISP's can start doing site caching for popular pages providing lower latencies for popular sites and pages.

you gotta admit, servers on the moon would be the perfect off-site backup.

Well, not quite perfect, you'd actually need a "server got hit by a meteor" contingency plan.

yeah, but you get rid of all those nasty weather-related contingency plans.

Until a solar storm rolls through.

In X hundred years people will have their real time mirrors of youtube/(representation-of-reality de jour), in their retina augmentations ;)

3000ms ping times were not unheard of in the early Starband satellite-broadband days.

I recall hearing Ted Nelson talk about his infamous Xanadu hypertext system some 20 years ago. At one point in his description, he quite seriously warned that "the system will be somewhat less responsive from the outer reaches of the solar system."

I gotta admit, the guy knew how to dream big.

That's 3 times faster then my Hyatt here.

By the time we have to worry about moon latency, I'm pretty sure all hosting companies will include a moon mirror.

How expensive could it be to mirror the entire internet? This is the future we're talking about.

...and the race begins to start the first interplanetary hosting company.

When is it cheaper to 'sneaker send into space' on a DVD, than transferring bits?

I doubt it will ever be cheaper. Faster perhaps, but not cheaper. Sending energy around space isn't all that expensive.

Apparently you can get stuff into space for as little as USD$400/kg [1] and a plain DVD weighs 16g [2]. Naïvely assuming ‘space’ to simply mean ‘Kármán line’ and discounting the cost of actually landing on the Moon, I estimate USD$24600 to get the DVD the average distance of the Moon [3].

[1]: http://en.wikipedia.org/wiki/Space_elevator_economics#Costs_... [2]: http://wiki.answers.com/Q/How_much_does_a_DVD_weigh [3]: http://www.wolframalpha.com/input/?i=%28avg%28363104%2C+4056...

Moon Wal-Mart will probably sell DVDs as a loss leader at around $24,588, putting all those poor mom-and-pop Moon stores out of business.

While the earth-to-moon latency might be high, the throughput would probably still be decent - I don't think sending DVDs would be cheaper.

The silver lining: moon computers won't need expensive cooling systems.

Actually, they will. Vacuums don't do shit for getting rid of heat. If anything, they'll need MORE expensive cooling systems.

And don't forget all the extra shielding you need to protect circuitry from solar winds.

Or you could use Athlon Duron processors.

In that case, the solar winds would need extra circuitry to protect it from the heat of the processor.

maybe they'll be used to heat the moon base.

You're saying the dark side of the moon isn't cold?

Or only that there is no way for the heat to move away once created?

I'm saying that you need something to transfer heat to in order to cool an object. In a vacuum, there is nothing to transfer that heat to - it doesn't just "bleed off into space" or anything.

And the dark side of the moon being cold has nothing to do with it being in a vacuum, I don't even see where you get that.

Reply on news.ycombinator.com