79 comments

[ 3.9 ms ] story [ 127 ms ] thread
Absolutely gorgeous.

I love it when these publishers get creative with the medium.

I agree that it's pretty and the medium really conveys the "cloudy" feeling with the text materializing from whiteness, but I think this is a bad thing for 2 reasons.

1. One of the cores of my cpu is at 100% while viewing the page.

2. It forces others to do the same kind of stuff for people to view it. I think every piece should be on the same level in terms of design. That way the pieces that are better written are going to be recognized more instead of being recognized for flashy javascript.

Of course, you cannot ban people from doing flashy things, and if it wasn't for flashy content, graphic novels wouldn't be a thing. Nevertheless, graphic novels are in their own section in a bookstore.

I think it's inevitable given that all news sites compete for traffic. The NY Times may take the high road, but they still need to compete.

If it weren't for the presentation, would you have seen this article?

> The NY Times may take the high road, but they still need to compete.

I find this argument intriguing and humorous given that the NYTimes itself set off this article design arms race with their Snowfall piece in late 2012.

Unlike the grandparent however, I have come to enjoy the kinds of experiences it has inspired. Using the web to render pure text strikes me as somewhat equivalent to filming a movie with just dialogue and no music.

http://www.nytimes.com/projects/2012/snow-fall/#/?part=tunne...

I'm not terribly impressed with the white (#FFFFFF) text on white (#DAECF0) background before I scroll down sufficiently either. I was restoring to text highlighting.
I think this is a bug where they didn't test large browser window sizes. At my laptop's maximum window height and my browser's default zoom level, the background doesn't get lighter than #BDDAE1 at the bottom before it transitions, but if I zoom out I can see it fade into unreadability.
Ahh, good call - my monitors are absurd. I tried it first on a 2048x1152... but I have that thing rotated sideways, so it's about twice the height of a typical 1080p monitor. Didn't fare much better on my normally rotated 2560x1440, but even then I nearly reach the bottom of the gradient before text colors change.

I guess I should just count myself lucky I haven't set up a rotated 4K monitor yet.

That's a similar mindset to people who think books are inherently better than movies/TV/games.

I prefer to judge a piece of content on its quality, not its medium.

Maybe the CPU being at 100% is intentional, so that it simulates the roar of the engines.
> 1. One of the cores of my cpu is at 100% while viewing the page.

Interesting. What browser/OS? On Safari on my Mac, the worst it gets is 20% of one core when one of the top/bottom animations are in view, but it drops to nothing when there's just text or one of the other, non-layered animations. And that's on a Retina display

Chromium and Firefox on Ubuntu.
I found this really well written.. captures the magic and puts you in the seat.
WOW - That must have taken a lot of extra time to polish, but totally worth it I hope - I definitely fell into the story with enjoyment. A new generation of writing.
The writing and the interactivity go beautifully hand-in-hand. I hope NYTimes continues doing this type of work.
And this is probably the worst one they've done. The Avalanche at Tunnel Creek was far, far better.
The writing is beautiful. The graphics are really well done, but just don't hold the same magic.
I'm really interested if anyone knows how the Times is planning to keep this type of content available on long time scales. They can realistically assume that twenty, fifty, a hundred years out people will be reading these articles.

These articles are a bit different from maintaining a text archive.

I imagine the interactive elements may not render correctly even just 5 years from now.
What interactive elements? All I saw as part of the content were static images, and newspapers have had those for a long time. There were some other illustrations and animations, kind of like the way some children's books have them interspersed among the text, but none of those were interactive either.
Another poster also referred to the animations as interactive... I guess some people think of animations as interactive.
I see animations that are driven by scrolling. Which you can control and scrub backwards and forwards again. So it's got an interactive element.
The Times broadly refers to this kind of storytelling as 'interactive' judging by this 2014 showcase: http://www.nytimes.com/interactive/2014/12/29/us/year-in-int...

I guess the underlying question is how they plan to support these web technologies into the future; what's their long-term technical debt in that sense?

Given that past coders for the NYTimes have included the creator of CoffeScript/Backbone and the current maintainer of ClojureScript/Om, it seems (at least anecdotally) that whomever is in charge of staffing the dev team is doing a decent job.

Of course, some of the worst technical debt I've had to deal with came from "world-class developers", so this is no guarantee that they won't collapse under the weight of their own code at some point...but I think the odds are at least in their favor.

(comment deleted)
Why not? It's not like web browsers habitually break compatibility, with some notable exceptions (removal of Flash and Java)... I guess it could break as part of a CMS refresh.
The chance that this will break because of a CMS update, a new paywall, site-wide Javascript updates is far higher than because of a web browser upgrade.
I read it through a reader that uses Readability, so it was plain text. I didn't know anything was missing until I read your comment.

That said, the text was a little flowery and aimless. In retrospect it feels like not was written to support the animations and not the other way around. If and when the format becomes more mainstream then it probably won't hold up as well in the future. Kind of like early 3D films.

There's even a substantial difference between mobile and desktop - and I didn't even know I was missing out. Some of the change is good (air traffic areas are labeled by their designators vs names on mobile), others are just missing which is a pity.
…it's adapted from a book.
The Times does have plaintext archives of pretty much everything - aside from anything else, it's how they make the print edition.
I can speak a little bit to that question.

We do, indeed, hope to keep interactive articles working and available over the long term. But you can't always predict the technical choices of the future web.

For example, there are plenty of past Times graphics done with Flash, which may not be easily viewable within the next 5 years.

On the other hand -- one of the biggest culprits is simply using external APIs as part of a graphic. APIs shut down, change their terms of use (Google Maps), or have data updated in ways that break the graphic. For example, as we just noticed the other day, this 2011 graphic used to show Joplin, Missouri before and after the tornado hit:

http://www.nytimes.com/interactive/2011/05/27/us/joplin-pano...

Now, the "before" panoramas are actually "way after" shots. We'll need to go back and fix it somehow.

But it's not all Sisyphean. The standard operating procedure for any interactive article page is to live somewhat inside and somewhat outside of the CMS -- with the page's own baked-out HTML base, and individual copies of any JS libraries and CSS it needs (apart from the super common ones). With a little foresight, this should be resistant to breakage from future site redesigns, even if the page continues to look a little "old" ten years from now.

As long as browsers in 2025 understand JavaScript from 2015 (likely), I'm optimistic.

"Always bet on Javascript" - Brendan Eich
Thanks for your comment, that's exactly the kind of answer I was looking for :)
I wonder why they keep using local & global air pressure-based altitude measurements instead of GPS. Is it just inertia, or does GPS have serious technical limitations when used in an airliner?

I can understand relying on airspeed instead of GPS-measured groundspeed, because airspeed is important. But I can't think of a justification for not using true altitudes, especially during takeoff and landing. What happens if the pilots punch in a wrong number for the destination air pressure?

I wonder how robust the GPS system is. Could an unexpected space phenomenon bring it down? What would happen if a rogue nation started shooting down the satellites?

This isn't a real answer. I hope someone with a real answer comes along!

GPS satellites somewhat regularly go in and out of service for various reasons. There are more satellites in the constellation than strictly necessary in order to compensate for this.

Information on this is distributed to pilots (in the US) via NOTAMs, and you can use this information to decide whether it's safe to use GPS navigation for your flight.

Practically speaking though, aviation GPSs have a function called RAIM that continuously monitors the integrity of the position information and will inform you on a loss of precision. I believe most GPSs also allow predicting RAIM along your time and route of flight to make sure you'll have integrity from takeoff to landing.

> I wonder how robust the GPS system is. Could an unexpected space phenomenon bring it down? What would happen if a rogue nation started shooting down the satellites?

GPS is not the only system. We currently have about 2.7 global satellite navigation constellations available for use.

The American GPS and the Russian GLONASS are both complete. The Chinese Beidou system is about 40% complete. The European Galileo system is about 30% complete. Each system also has about 20% redundancy in the form of on-orbit spares.

Dual-system GPS/GLONASS receivers are now nearly everywhere. I was just looking at a phone that has a triple-system GPS/GLONASS/Beidou receiver. In a few years, quad-system receivers will be commonplace.

A rogue nation would have to shoot down about 1.9 constellations, or about 60 satellites, before we drop below full coverage.

The American, Chinese, and European systems have the same orbital inclination. If a gap opened, any satellite from any of these three systems could theoretically fill that gap. It would take some time to reposition the satellites, though.

There is also additional redundancy in the eastern hemisphere. India and Japan have satellites that provide regional coverage. Part of the Chinese system also operates as a regional constellation.

At this point, the rogue state is at war with the US, Russia, China, the EU, India, and Japan. They might as well just surrender.

It would take only one nuclear device being "detonated" in low orbit, to wipe out nearly all of the satellites in a large area of that space. The resulting EMP may wreck quite a few ground-based systems too, IIRC from a book I read recently. Can't recall the name, though...

In any case, disrupting satellite-based navigation is the simplest kind of rocket science: a missile that just has to go up and then detonate, no aiming required.

> It would take only one nuclear device being "detonated" in low orbit, to wipe out nearly all of the satellites in a large area of that space. The resulting EMP may wreck quite a few ground-based systems too, IIRC from a book I read recently. Can't recall the name, though...

A nuclear weapon detonated in low earth orbit would not destroy a single global navigation satellite. They are not in low earth orbit.

A nuclear weapon detonated in medium earth orbit would wipe out between 0 and 1 navigation satellite, depending on how close you get. The satellites just aren't that close together.

An EMP would destroy electronics on the ground, but this will be a systemic effect. Satellite navigation would be collateral damage. When the ground-based electronics are replaced, though, the satellites will still be working.

> In any case, disrupting satellite-based navigation is the simplest kind of rocket science: a missile that just has to go up and then detonate, no aiming required.

Nuclear weapons are powerful, but they still have to obey the laws of physics. If you get far enough away, the inverse-squared law makes even a nuclear explosion look like a firecracker.

A 1 megaton nuclear weapon would have to be detonated within half a kilometer to damage a satellite physically: http://www.projectrho.com/public_html/rocket/spacegunconvent...

The shielding needed to protect a satellite against EMP only adds 5% to the cost of the satellite: http://fas.org/spp/starwars/congress/1997_h/h970716u.htm

So basically, it's going to take one nuclear warhead to destroy one navigation satellite. If a rogue nation has 60 nuclear warheads, it's hardly going to waste them to take out 60 navigation satellites.

You'd need to be able to ensure that everyone is using GPS altitudes before requiring it, and there are some pretty ancient planes still flying out there. As far as I know, ATC radar doesn't actually measure altitude but rather relies on the plane's transponder to report it.If you have some mix of baro and GPS altitudes, you can no longer safely ensure separation.

There should be enough checks for setting the proper baro setting - both pilots check the setting individually when changing it and it's reported to them a number of times throughout descent. Even if they do manage to get it wrong through all that, the GPWS should notify them if they get too close to any obstacles.

And not everyone uses GPS (GLONASS is a competing system used by some parts of the world, specifically developed by the USSR to not have to rely on a USA based system).
I think the biggest hurdle is lack of precision. 200 feet isn't much error in the horizontal realm but is a big deal vertically.

There are ways to supplement the precision of GPS altitude with ground-based stations (WAAS), and this is indeed used to construct GPS-based instrument approach procedures which are comparable in accuracy to traditional means (ILS). But these require ground-based transmitters; impractical to implement worldwide.

The most unassailable answer is that the procedures were worked out before GPS equipped airplanes were in the fleet, and even today, GPS is not required equipment for flight in the flight levels (and many older jets are not equipped, especially older freighters).

Other reasons are: the transponders that ATC (and other aircraft) interrogate report pressure altitude (uncorrected, meaning using 29.92 inHg as the standard).

A sensitive altimeter can be had for a few hundred dollars and requires no electrical power (unless lighted) and basically no service throughout its life.

To your last point, anywhere it matters (below the flight levels), pilots are using a local altimeter setting on the altimeters, often from the ATIS (automated terminal["airport"] information system) or from ATC. Of course, for takeoff, it doesn't matter. You're generally climbing at max available power or are doing a reduced power takeoff in aircraft so certified and on runways without obstacles to worry about.

On approach, you're typically following a precision approach path provided by a radio signal (ILS includes lateral and vertical guidance beams) or that is GPS-based (where you have similar navigation performance levels provided by WAAS augmentation). There are non-precision approaches where the altimeter setting is more critical, but I can't recall the last time I had to fly one of those for other than training purposes.

Even the newest airliners of today still triangulate their positions off of radio beacons (VORs) as a backup to the GPS system and also contain an inertial system. GPS being the most accurate, it's normally the primary navigation system, however, I've flown across the country, beacon to beacon, with the GPS inoperative, which some airliners still do from time to time.

In the southwestern US, especially, GPS often becomes unusable due to military activity. It isn't uncommon for an airliner to experience a loss of GPS signal mid-flight when operating near military exercises. It's just the nature of our system.

It was a good article. I enjoyed the allusions to ships on a sea, and moment before landing described as an unanswered question. I'm wondering if this will be passed along much in the pilot community. The most popular 'pilot life' creation in recent memory seems to be this Lego parody - https://www.youtube.com/watch?v=rNxz2hhSXuY.

The technical limitation is the basic geometry of GPS. Satellites are at ~20,000km, aircraft at ~10km max. And you want 0.0003km resolution in altitude. We barely even have 0.003km in the lat/long plane, and it's only that good because of fancy DSP algorithms which lean on special modeling of atmosphere and the satellites themselves - I.e. assumptions about a 3rd-party system, wholly inappropriate for safety critical stuff. In fact, cruising altitude (~"full span" indication) is roughly the magnitude of the specs for error in lat/lang back when GPS was first commissioned (IIRC).
All the other replies are true, but the main reason is that airplanes fly a constant pressure altitude because it's important performance wise, while flying a constant geografical altitude is almost meaning less(except for obstacle avoidance.

Pressure altitude is fundamental for the performance of airplanes because you fly in a known estable layer of the atmosphere that keeps your drag, lift and thrust almost constant (there are temperature changes but it takes docens of miles to change a single degree).

Also if you select the standart atmosphere you know you are flying the same layer system of all the other airplanes in the air, with a fully autonomous, relatively cheap, and non electrical device. Also it's easy to calibrate on your own if you want to fly in a local airport and have no access to a weather report. It's hard to beat a conventional altimeter.

GPS altitude is considerably poorer than GPS lat/lon. There's a good chance that your GPS altitude could be 10-20 metres off what it should be, which is enough to be dangerous if you're using it as a landing aid.

"True" altitude requires you to both have an excellent fix on the satellites and have a very, very good reference geodetic model (which you probably don't). Also depending on the quality of your fix, you'll have different altitudes to the rest of the people around you.

It makes more sense to have two pilots independently check the pressure settings (and that's why you have a checklist).

My dad was a navigator on a B-17. He'd tell me that he had to scrape the frost off of the dome with his fingernails so he could get a fix on a star while flying across the Atlantic. Crossing the Atlantic at the time was at the limit of the B-17's range, so messing up the navigation means you disappeared. Nobody dreamed of GPS back then :-)
Some operational C-130s still have a little sextant port for shooting a star but only a few of the old guys would still even know how it works.

The GPS most of them use is almost as old, a $20 Garmin handheld could navigate circles around it.

> a $20 Garmin handheld could navigate circles around it

Probably not, as it will refuse to work at both the altitude and speed of an airplane ;-)

EDIT: I stand corrected. Neither the speed nor the altitude will be a problem.

> Probably not, as it will refuse to work at both the altitude and speed of an airplane ;-)

Well sub in the garmin with one of the other low cost consumer units that work at those altitude and speeds then!

The GPS altitude limit is 18km which is 60,000ft. Normal cruising altitude is 35,000ft.

It's also trivial to obtain unrestricted GPS devices.

It is really interesting how the bottom part of the text fades away progressively. It elicits an emotive reaction to want to scroll further down. Well done!

I can't spot any css, fixed divs etc. Thoughts on how that is being done?

(comment deleted)
On the other hand, it slightly annoyed me, I guess for a very specific reason, which I'll state in case anyone finds it interesting. I'm nearsighted and wear glasses; as I often do, I was using my laptop with my glasses off, because taking them off moves my default focus point closer and thus lets me focus on nearby objects with less eye strain. However, the result goes a bit too far in the opposite direction and I need to keep the laptop closer to my face than normal to be able to focus on the screen, which means the high-detail center of my field of view covers less of the screen than typical and I have to move my head slightly to shift it. On top of that, when I started reading the article, in an unconscious attempt to focus on the text, I had leaned forward to move my head even closer... And so I had been reading using solely a paragraph or two of space, which happened to be at the bottom of the screen (gently scrolling the trackpad to continue - this is way more precise than using a mouse scroll wheel, by the way)... for the first part of the article; then that space suddenly became half greyed out, and the required head movement to switch to reading higher up was unusual enough to bubble its way into conscious thought.

Oh, and I guess it doesn't help that a fade to white at the bottom of the screen is also what the NYTimes paywall does when you've run out of free articles - I have a paid subscription, but sometimes I get logged out. This has happened often enough to cause me to form a mental association between the bottom of a NYTimes article suddenly greying out and the annoying act of logging in. Pretty pedantic complaint, I know.

I wonder if I'm the only dev who completely ignored the presentation and just enjoyed the story!
It looks great, but be honest: did you really read through the whole story? I got bored pretty quick. The text just lost me..and it seemed to be a neverending css showpiece, which is fine, but I don't think you can disguise that as an interesting news piece for too many scrolls.
Exactly right. I just ended up skimming it and then skipping to the end.
On one hand, I'm annoyed because I know that this book excerpt appears in the Times as part of a marketing campaign.

On the other hand, this was probably the greatest thing I've read in a long time, so I guess bring on the paid content!

I fly about 100,000 miles a year, but this article absolutely brought the wonder of flight, and, in conjunction with the in-situ animation, just pulled me in. I couldn't stop reading for a second. Absolutely wonderful, and brought a lot of the mystique and thrill of flying back. I hope the NYT does more articles like this.
Fabulous article, elegant design. Nice touch with the NYT logo and a little hiring message console.log'ed:

"NYTimes.com: All the code that's fit to printf()"

I started studying to fly when I was 16. Judge preemptively revoked my driver's license, so I figured I'd fly just to show him how petty he was. Up til then, it had just been JetFighter II for me.

This article nails it in the opening. I don't remember a whole lot about my first flights (well other than almost hitting a tree - JF2 was mostly carrier landings so I had bad habits). But the feeling of pushing the throttle, watching the marks turn to steaks, and the feeling of holy-shit-I'm-flying -- it's amazing. I recommend getting a first lesson (should be cheap-ish) just to feel that. In nearly 20 years, it's one of the top physical experiences I recall with exhilaration.

Ditto. Every time I take off and I'm at the controls of a plane, it is pure wonderment. Best feeling in the world. One day I will buy but for now I rent. I recommend it to everyone to get their license or even a sports license.
have they developed a new CMS to produce articles like these ? reminded me of the Apple watch review article by The Verge
More interesting waypoints.

The SSTIK1 Departure from KSFO has a few good ones: UTOOB, NTELL, AYOOH, EBAYE. http://flightaware.com/resources/airport/KSFO/DP/SSTIK+ONE+(...

One that always made us chuckle was a standard routing from London over Belgium, the the Biggin Hill VOR, BIG, the KONAN waypoint, then the Koksy VOR, so the route was BIG KONAN KOK.

The eyebrow raiser was called DROWN, and its right between Cuba and the Southern tip of Florida. Not accidental.

Little Rock Airport has the ICAO code KLIT. There's heaps of good ones.

I used to dispatch international flights. The author pulled back the veil on some esoteric aviation knowledge and still managed to make a good story of it. Good form.

I'm not a super-fan of the jingoistically named waypoints.
its not worse than the naming many give to their languages, tools, and such, in the tech community.

light humor sometimes invades work and for that it serves a purpose. I also tend to think a little humor or local flavor improve memorization

My favourite is the "Looney Tunes Approach" at Portsmouth NH (KPSM): ITAWT, ITAWA, PUDYE, TTATT and IDEED.
Was looking up some more info on Maastricht Upper Area Control Center (MUAC) and found [1] this image of a controller in front of a terminal with a Tux sticker on it: http://i.imgur.com/VQFn2hM.jpg

Does anyone have an explanation for that? Are they using FOSS at these centers?

EDIT: Apparently the stickers aren't on every terminal but at least on more than one, as can be seen in this picture: https://www.flickr.com/photos/eurocontrol/5201407322/in/albu...

[1]: http://www.eurocontrol.int/muac