Hacker News new | past | comments | ask | show | jobs | submit login
Dreamliner spells out "787" & Boeing logo over US during test flight (flightaware.com)
315 points by ChrisArchitect on Feb 10, 2012 | hide | past | web | favorite | 80 comments

Similar attempts:

Gulfstream V: http://flightaware.com/live/flight/GLF17/history/20070206/15...

Cessna Logo: http://flightaware.com/live/flight/N750CX/history/20080307/1...

Boeing 747: http://flightaware.com/live/flight/BOE523/history/20110802/1...

Also, a thread from one of the pilots: http://flightaware.com/squawks/view/1/7_days/popular/24611/B...

Interestingly, this required a lot of work on our side to pull this off. Since the route field on flights plans expects short input and this flight's route is dramatically longer, it caused an FAA system to split the flight arbitrarily into different legs.

(Disclaimer: I work for FlightAware)

I'd just like to say: Thanks for running such an awesome site!

I use FlightAware to hack airlines by trying to locate the incoming flight that is to be my outgoing plane. Any plane landing ~60 minutes before my departure whose make and model matches my flight is a suspect. I can usually predict whether or not we'll have an on-time departure long before announcements are made.

On some flights, we actually know what the corresponding inbound flight is. If we know, we show a small "track incoming flight" link and consider it in our delay prediction algorithms.

On time prediction, my friend, is a profitable business. Just in case you decide to quit your current job :)

Have you seen FlightCaster?

~60 minutes before my departure

60 minutes? Wow. Ryanair here in Europe does a 25 minute turn around. Often you're through the boarding gate and queueing up before the inbound flight has even landed.

> I work for FlightAware

Cool! Say, are you guys still running on Apache Rivet, just out of personal curiosity?

Yes, we do (we're about the only serious production site running Rivet left). Our codebase is almost entirely in Tcl (sigh). Apart from that, our core algorithms are in C and we run a pretty serious production Node.js app.

Tcl's pretty cool, and I think it could have kept going interesting places if things hadn't fallen apart in some critical ways. Especially back in the day when the competition was stuff like PHP or Java. But it's no fun to be working with something that's not headed anywhere.

It's great to see someone using Rivet, though. Damon Courtney and I wrote that something like 10 years ago, with the occasional input from Karl. I wish I had half the business sense that Karl seems to, he seems to come up with a lot of good products that also make money.

> Tcl's pretty cool, and I think it could have kept going interesting places if things hadn't fallen apart in some critical ways.

Hey David, nice to see you again :)

With respect (as always) I don't get what you're talking about. You've written over the years that Tcl suffered from a marketing problem. No question. The rest of your anti(?) Tcl writings seem to make allusions to problems, but never really express what the problem _is_. I'd love to spend some with discussing this with you (email perhaps, or in quiet irc space?).

When I describe Tcl, I start similarly to you (Tcl's pretty cool...) and build on that, rather than detracting from it. Tcl is _fun_ like nothing else I've played with, has a brilliant C API, is incredibly dynamic, etc., etc.

Even despite the apparent lack of popularity, it's _still_ everywhere. Expect for system administation and job scheduling, etc,. Tcl in Cisco iOS, Tcl in F5 Network appliances, A10 appliances, driving Git gitk, embedded in Python to drive Tkinter, and a scripting/GUI interface for R, CSound, Vtk, ns-2 network simulator...

Tcl needs positive stories and it's successes pointed out, not "It could have been cool, but it's not". Regardless of where it sits in a popularity contest, Tcl is awesome, time-tested, rock solid... and _still_ improving.



Didn't TCL not get constant-time array lookup until like version 8?

It's had "lists" (arrays) since... forever, as well as "arrays" (hash tables).

Also appearing in version 8.1 (in 1999) were Unicode and good support for threads, stuff that systems like Ruby only got recently.

Yes, but iirc array lookup was not constant time.

Hash lookup is constant time. The constant is merely higher than for a multiplication-and-addition-based vector lookup.

Just wondering if you guys are setting the default language of the site based on the browser's location. The site opens up in Hindi as I'm accessing it from Bangalore.

We do, unless you have (a) changed manually your language, in which case we save a cookie or (b) you're logged in and changed your language in settings. Also, (b) takes priority over (a).

I love FlightAware, thanks for making it awesome.

I actually posted this about 16 hours ago (http://news.ycombinator.com/item?id=3574619) and I point this out because it's always fascinating to me when the "right" time to post something on different sites is.

For example, I believe days start at 8pm PST on Flickr (can anyone correct me if I'm mistaken) and there's a power-user perspective that will wait to post things so that they can get a leg up on the "interestingness" algorithm.

Anyhow, this post makes me smile even the second time around.

I think part of it is the title also. "787 plane draws a 7-8-7" doesn't tell enough to interest me, while "Dreamliner spells out "787" & Boeing logo over US during test flight" elicits an immediate "Oh, cool!"

Timing definitely has something to do with it, but also consider the headline.

Yours was "787 spells 7-8-7", which is not as simple to understand as "Dreamliner spells out...". Instantly, this headline catches enough of my attention to click on the link.

For headline copy, sometimes it will help to search Digg/Reddit and see which keywords people have already clicked on the most. This applies to email subjects, website CTA, etc.

(doh, wrote a long reply with thoughts a few hours ago from Hacker News Beta mobile app and it didn't post...fail)

sorry man, it was just the link I got. I'm all for first poster credit and respect, just sometimes same content can be generated from multiple links etc and stuff gets messed up. Annoying but you know how it goes.

On the time of posting thing...I know someone created a site to determine the optimal posting time for HN.... I have never used it tho! It's just naturally there'll be hotspots in the day based on simply where the majority of HN readers are from and the waves of social network activity.

Anyways, a good link indeed. A fun Friday afternoon share -- Great that the FlightAware guys are watching this and contributing too.

ah, the 'tool' for optimal HN Pickup. Still interesting when you look at it. http://hnpickup.appspot.com/

Also, re: titles -- yeah, descriptive titles always -- I try to do that, since I appreciate it when others do

My Flickr daily stats reset at 4pm pst, not sure if that matches the "interestingness" day, but it seems likely.

Which is midnight, GMT.

The test pilots are basically told what part of the world and for how long they have to fly; when they turn in a plan like this basically for the sake of cuteness, that's a lot of work picking waypoints.

This pleases me.

Question for people with more aviation knowledge than me: Don't games like this make life hard for air traffic controllers? 35000' is right in the middle of normal airspace; isn't it much easier if planes are flying along standard routes between airports?

Altitude is determined by direction in alternating chunks of 1000ft. So, say, southbound traffic flies at 29000, then 30000 is westbound, 31k north, 32k east, and then 33 is southbound again.

So at cruising altitudes every airspace can accommodate travel in any direction... the controllers along the way probably didn't even realize that anything out of the ordinary was happening except for talking to a Boeing callsign (as opposed to the call sign of one of the major airlines that typically operate flights at this altitude, Boeing operates its own "airline" in the regulatory sense just for test flights).

Given the length of that route (and all those arbitrary GPS waypoints!) I'd think ATC would catch on pretty quickly.

ATCs along the way don't necessarily need to get the full picture. Each one is responsible for a zone. They have planes about to enter their zone, they route them/land them and then the plane moves out. Not saying they can't easily get the flight plan and figure it out, but they probably wouldn't normally do it.

But no matter how small the zone is, the point where the plane is to make a 90/180/270 degree turn is only in one zone. That means the ATC responsible for that zone would see it. Are normal planes really doing 90 degree turns?

I haven't looked at the route that Flightaware provides yet, but there are a LOT of defined navigational fixes and standard routes between these fixes. Its possible the crew simply assembled a series of fixes (and standard routes) to get the desired shape.

Enroute charts tend to look a lot more like a US highway map than an Interstate system map.

This flight got up to 42,000 feet if you look at the elevation graph. I didn't know planes flew that high.

<http://www.messybeast.com/dragonqueen/cockpit-chatter.htm...; (Flight level is x10000, so FL600 is 60,000 feet)

SR-71 Blackbird was crossing the control-zone of London Control. Evidently the controller didn't know the service ceiling of this aircraft is around 30,000 metres (yes metres, not feet!). There are so many variations on this item that the original event has evidently become part of aviation folklore or joke-lore.

Pilot: "Radar, Good Day, Airforce Blackbird, request FL 600(!)"

Controller (amused): "Sir, if you can reach, you are cleared FL 600"

Pilot: "US Air Force Blackbird, leaving FL 800, descending Level 600..."

So many good stories about that bird.

One of my favorites:


"There were a lot of things we couldn't do in an SR-71, but we were the fastest guys on the block and loved reminding our fellow aviators of this fact. People often asked us if, because of this fact, it was fun to fly the jet. Fun would not be the first word I would use to describe flying this plane. Intense, maybe. Even cerebral. But there was one day in our Sled experience when we would have to say that it was pure fun to be the fastest guys out there, at least for a moment. ..."

That's high for a large body airliner, but some smaller business jets can go higher because they have better aerodynamics and engines. Service ceiling figures from wikipedia:

  Gulfstream 550 and Dassault Falcon 7x: 51k

  F14, F18, B52, and B-2: 50k
  F15, F16, F35, B-1B, Su-47: 60k
  F22, Typhoon, and PAK FA: 65k
  U-2: 70k
  SR71: 85k

  A380, A350: 43k
  A340, A330: 41k
  A320: 39k

  Boeing 787, 777: 43k
  Boeing 757, 727: 42k
  Boeing 747-400: 41k
  Boeing 737: 35k

The 85k for the SR-71 is the unclassified information that has been made available. Rumor has it they hit 100k+ with the SR-71. Figures on actual ceiling for the SR-71 flights that were made have never been made public.

> I didn't know planes flew that high.

Well, it is a test flight of a brand-new plane. Also, some "planes" can go as high as they want in the atmosphere, even up to space (kinda).

I don't know what you mean by "planes." All airfoil based aircraft have a limitation on how high they can fly before the air gets too thin for their wings to generate sufficient lift (known as their "operating ceiling").

The only exception to this I know of would be a craft with enough thrust to use power as the source of lift (like the space shuttle).

My flights between Taiwan and California in the last ~18 months were usually around 40,000 feet give or take a bit in 747s and 777s.

The entire path is communicated to ATC in advance when the flight plan is filed, so they can foresee the route on screen when directing other traffic. Also, much of the flight was in the evening and late night, so traffic controlling load would have been lower for them.

Reminds me of the 'biggest drawing in the world' viral image that went around a while back, which later turned out to be a hoax: http://www.wired.com/gadgetlab/2008/05/artist-says-he/

Pretty cool that someone actually did it though!

A 19 hour flight? How was this even possible without running out of fuel?

An empty plane is pretty light so it burns less fuel, AND because they don't have passengers and cargo counting against their MTOW (maximum takeoff weight) they can take off with more fuel.

Most planes don't leave topped-off - they leave with the right amount of fuel to hit the maximum takeoff weight given that day's load.

Not to mention that extra fuel (20 hours of fuel on a New York to Chicago flight) becomes "cargo". No point in burning the fuel (and corresponding dollars) to ship fuel the most expensive way possible to an airport that already has plenty.

Newark - Singapore (Singapore Airlines Flight 20) is regularly scheduled for 18hrs and 50mins.

Wikipedia has a good list of long, regularly scheduled, non-stop flights: http://en.wikipedia.org/wiki/Non-stop_flight#Currently_sched...

EDIT: Flight is SQ21, not SQ20

19 hours on a plane must be quite a sufferfest. I do Europe->Oregon every year or two and 12 hours is already agonizingly slow. About the time you're thoroughly sick of the movies, reading, drinking or anything else, you realize you still have another 4 hours to go.

When Quantas had a San Francisco -> Sydney non-stop it was 14+ hrs. It was the longest flight I had been on up to that point. Fortunately I was in business class and could stretch out, but I remember watching two movies, taking an 8hr "nap", and then STILL having another 2hrs of flying to do. A looooooooooooong way to go.

Business class is a lifesaver for long-haul flights. I can't sleep unless I can (mostly) lie down; noise and motion aren't a problem, but I need a seat that reclines if the flight is going to cross more than four or so time zones.

Economy seats for most airlines have something like an incredibly generous 5" of seat travel, which is why I've caught up on all the movies that I didn't watch when I was younger.

Oh, and if I start reading (which includes hacking), I get motion sick. It's hell.

I'm genuinely envious of people that can sit and work in economy. It would be fantastic to be able to get some hacking in...

Quick tip: get yourself some prescription sleeping pills, and take one every long flight you take (and only for flights—they're dangerous drugs!) They make the flight a breeze.

If you're going to do this (and I've done it myself), be aware that it may increase the risk of blot clots (deep vein thrombosis). You're supposed to try to get up and move around every couple of hours on a long flight, and taking an Ambien or the like may mean you'll go 6-8 hours (or more) without doing that. Plus, you won't be drinking any water, which isn't healthy.

One of the best ways I've found to both increase hours slept and reduce overall stress on long flights is to buy some cheap foam earplugs and wear them the whole time. (Noise reducing headphones are great but they're bulkier, more expensive and have to be turned off during takeoff and landings.) I used to fly red-eyes fairly regularly and having a bag of those bright orange foam earplugs in the pocket of your work bag was a lifesaver.

I found what also works are the Etymotic headsets, if you are the type of person that doesn't mind the in-ear style headsets. During takeoff and landing, I leave the plug dangling from my hand and obviously visible so the flight attendants don't bother me about turning off my $music_player. One less thing to keep track of while traveling, and fills multiple purposes.

DVT is no joke on these marathon flights, glad you pointed it out to others in this thread.

Yeah, I actually use Ultimate Ears (if I remember to put them in my bag.) I've found that there are some advantages to earplugs, though - they're so cheap you can leave a dozen in your work bag (and give a pair to your travel companion if they need them) and there's no wire to worry about when you're just trying to sleep.

The risk probably depends on your age. I've flown from LAX to Hong Kong and back in business class before, and the flights leave after midnight local time. I found it no trouble to sleep for 12 hours without taking any drugs. No blood clots, no dehydration :)

Neither of us can really hope to assess the change in risk of something like DVT based on personal experience. We're talking about something that normally kills some tiny portion of travelers (1 in 100,000 passenger legs, maybe?)... if not moving around increases that risk 10x, you'd still have to do a few hundred thousand flights to know.

Edit: The annual risk of DVT apparently runs about 1 in 3,000 for the general population. So if you were a higher risk passenger (older) and flew really regularly, I could see it getting down towards say 1 in 100 annually. Even then, you'd still need decades of data to make any conclusion.

survivor bias. The ones who did have problems are not on HN to talk about it anymore :-(

(I'm speaking from experience, it happened to a friend - young, healthy - be careful).

OK, but how is lying in bed on a plane different from lying in bed on the ground? (If it's the difference in air pressure, wouldn't everyone in Denver and Mexico City be dead by now?)

For one thing, when you're lying down, your feet are generally level with your heart, which makes it easier to circulate blood through your legs. Also, you're going to change position naturally while sleeping in a bed, but you're much less likely to do so when jammed into an airplane seat. Plus most people don't spend more than 8-10 hours in bed at a time, whereas DVT is primarily a big concern on long flights stretching 12-20 hours.

haha! nailed it exactly! I've done that flight (Portland to Amsterdam) a few times and it's right around 3-4 hours left that I hit the "are we there yet" moment.

It arrives two calendar days after it leaves!

One of the pilots from this test flight even remarked, "As a matter of fact - we didn't even top off the fuel tanks."

The 787 is designed for this long of a trip. There are a lot of trips out there, like the central US to the Far East where a direct flight is amazingly useful and economical.

Massively reduced weight without passengers and cargo, maybe?

Can't wait for the green-team to have their arms up about wasting fuel to do this (even though it's a scheduled test flight)

It's great to see them do something interesting while they're already in the air burning time/fuel on the new engine testing. The test flight would have occurred for the same duration regardless of the path. They needed a flight that was about 19 hours long for extended operation certification purposes.

19 hours and not over water which constrains them quite a bit. If they're going to run laps around the continental United States, why not make it fun?

Even with multi-engine they can't run over water for the certification?

It's a test flight, so you want to keep near some sort of airport in case something goes wrong. So yeah, you could duck over the ocean, but you really don't want to be flying in areas that could be problematic for too long in an untested plane.

and here I was hoping they were stunt flying it.

I can't wait to fly in one of these. The wings bow like crazy, can't wait to see them out the huge windows.

WTF? Completely unreadable.

This seems like a waste of fuel for a stupid publicity stunt. Airplanes dump so much carbon dioxide into the atmosphere.

If it were just a publicity stunt, I might agree with you. But they have to use up the fuel in a test flight anyway. It's not like they can go up for 20 minutes and say "all good" and end the test.

They were going to be in the air for 19 hours as a test flight anyways. Who cares what the path looks like?

There's being environmentally conscious, and then there's being a green freak.

That's true, I was being a "green freak."

Would you prefer they tested it with passangers?

My comment was serious. They have to test the plane, why not fly around in a pattern that is interesting?

It's much more important to be environmentally conscious with the bigger stuff. One test flight isn't going to make the tiniest dent in worldwide carbon dioxide emissions. It's much more important to advocate for changes on an industry-wide and systemic level, rather than protesting one-off slightly wasteful novelties.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact