Traceroute from a Norwegian Airlines airplane | Hacker News Reader

Traceroute from a Norwegian Airlines airplane

gist.github.com

84 pointswatson32 comments

I can't excuse all the nodes in that traceroute, but the ones with a domain "direcpceu" are Hughes Communication satellite ISP nodes. Their satellites are geosynchronous, so, for two legs in each direction, those packets went a very long way before reaching the ground and coming back to the plane.

Thanks for that. I was wondering where the transition from air to land was. Makes sense they route through Satellites. It makes it more interesting because your packets are going from 39,000 feet to space then to land. Awesome

This is implemented over Ku-band by Row44...

http://new.row44.com/faq/

I started this trace-route within 30 seconds after the seatbelt sign was switched off (the wifi isn't enabled while taking off), so I imagine only a few people would have had the time to connect to the wifi at this point.

But in general, the wifi on one of these flights is very useless. I was hacking on a node.js project and used it to look up some ES5 related stuff on StackOverflow. I tried to post it to gist.github.com while in the air but wasn't even able to load the page (though google and SO loaded fine)

I also tried to comment on a post on Stack Overflow. The first 5 times I clicked the button, I got a timeout and had to try again. But in the end some lucky guy got a response on SO from 10km above Europe ;)

GoGo in-flight wireless in the USA is using a terrestrial network that piggybacks on cell towers. I've had 1.5 Mbit with ~400 mSec ping times, which is a lot more usable than the satellite systems.

GoGo uses their own equipment on ~400 AT&T towers across the US, networked together using AT&T's MPLS network.

Disclaimer: I interviewed at Aircell at one point.

Next time you get the chance to ask astronauts something (in the Netherlands we had that once with Andre Kuipers I think), ask them to do a traceroute from space!

I wonder if Chris Hadfield would be willing, he seems to be pretty engaged with us on the ground: http://twitter.com/cmdr_hadfield

Wow, look at those response times: impressive.

I've had a better experience from Norwegian planes than from some hotel WiFi networks.

and me. The Wifi is amazing considering you are at 39,000 ft.

...and it's free. I'm super impressed with Norwegian Airlines.

just to be clear - I am not saying that the Wifi was broadcast from/to 39,000 ft ;-) Obviously I am referring to the wifi in the cabin that connects to the Norwegian Air system...

I wonder if it would be practical to install upwards facing antennas on cell tower and then have a tracking dish on planes track these to cut out the satellite when over (populated) ground. If a line-of-sight microwave link could be established, it should be possible to have high-bandwidth low-latency connectivity in the air.

What is the effective beamwidth of a microwave antenna that needs to maintain a 50-100km link? Would the ground antennas need to be tracking as well?

Tracking aircraft is not practical unless you're going to use a phased array, and the problem you'll run into then is how large the cells are when you're pointing towards the sky, reducing spectrum separation.

Why is tracking aircraft impractical? In university I was involved in building an antenna that would track a LEO satellite (a bit crudely, though), with a pass taking about 15 minutes - that's pretty similar to a pass by a full-altitude aircraft. The main impracticality I can see is that you'll need an antenna pr. aircraft in range.

If you want to track a number of aircraft, and not require an antenna/dish per each or have resultant mechanical breakdowns, phased array antennas are the best solution.

Thanks for responding.

This is _exactly_ what Aircell does in the US. I'm not sure what spectra they're using, but the bandwidth would likely be quite variable, as planes with non-directional antennas will interact with one another over wide areas. Weather could also play a part, depending on the frequency.

This (Air-to-Ground) is what Gogo does, using EV-DO Rev A.

http://en.wikipedia.org/wiki/Gogo_Inflight_Internet

Last couple flights I've usually seen latency between 300ms and 3500ms, 8-10% packet loss. They also seem to use Squid proxy, which doesn't understand WebSockets.

A minor nitpick: I think the author means simply the airline known as "Norwegian", formally "Norwegian Air Shuttle ASA". "Norwegian Airlines" makes them sound like our main carrier, while they are in fact a (very popular and vastly successful) low cost, point-to-point-only, no-frills one.

Southwest apparently uses Row44 for in-flight WiFi. When it even works, it's absurdly slow. So slow that I'd gladly take a dial-up modem experience instead. On top of this, the company insists on slapping up a banner on the top of every webpage. If I click the "go away" button, I have to wait 30 seconds for the page to re-load. [Yes, I'm sure GreaseMonkey, etc. could rescue me] I fly on Southwest so often that I receive free WiFi, but I rarely use it. Even logging in is a multi-minute ordeal. I can't see how Row44 can generate enough revenue to justify its existence given the abysmal user experience.

Well Google will probably peer someday with direcpceu.com (Hughes)

I think it's using a satellite network in orbit just like people with rural internet?

http://en.wikipedia.org/wiki/PanAmSat

I think Louis CK responded to this best: http://www.youtube.com/watch?v=b3dYS7PcAG4

Am I missing something? Why is this on first page of HN. Not downplaying, I honestly don't know.

Same old reason. Members of HN voted up because they found it of interest.

Now, tell me, did you really honestly not know that? I bet you did ;)

I think GP is really asking "Why do people find this interesting?"

A better question is: "Who doesn't find tracing packets from 39,000 feet interesting?" =)

Not that I care, but do the people who are down voting me, care to explain why? I don't think my comment was derogatory or anything, I just failed to understand why this article was amusing, and wanted to know what makes it interesting. I am sure other people might have the same question, and would have benefitted from the answers.

This appeals to me because I love anything dealing with aviation, you might love something else that I find to be an absolute bore and vice-versa. Is there any point discussing why we like and dislike what we like and dislike?

I thought there was maybe something I was missing, e.g. the speeds being extraordinary or something. I guess people are too trigger happy when it comes to downvoting a comment.

Reply on news.ycombinator.com