Hacker News Reader

Comment by Fenn | Hacker News Reader

This. (Disclosure: I'm cofounder/CTO of Volantio). Airlines are very much currently burdened with a legacy that was cemented in sometime in the 50's.

It may be worth mentioning that airlines are perhaps less-evil than is apparent from the outside. Most of their baffling behaviour is (IMHO) largely due to lack of capabilities rather than malevolence.

Ultimately, they're given a large multi-dimensional optimisation problem in a marketplace with extreme illiquidity (eg: traditionally, seats can't be resold effectively once purchased).

The complexity is, frankly, staggering - the seminal paper on the topic is probably this: http://www.demarcken.org/carl/papers/ITA-software-travel-com...

We are taking the (very small) initial steps in alleviating some of the issues by introducing increased liquidity into the market, but the ultimate goal is helping shift the industry into a more dynamic place (perhaps as described by the parent comment).

View on HN

"Airlines are very much currently burdened with a legacy that was cemented in sometime in the 50's."

Personally, I find this common observation off base. Most airlines have pushed the legacy TPF layer into a sort of nosql data store. Both Sabre and Amadeus have been able to pull the business logic out of TPF.

Any sort of lagging technical progress seems more related to either thin margins, or the complexity you point out.

There's certainly room for innovation and improvement. I just hate the framing of the problem being tied to "old systems". It is mostly not true. Varies per airline, and res system, of course.

Sorry, I possibly stated that poorly - what I meant was: The current state of duopolistic airlines technology suppliers means that there may not be sufficient motiviation to evolve the capabilities of the old systems.

Case in point: Regardless of whether Sabre/Amadeus have replaced their legacy layer with some sort of nosql, I can absolutely guarantee you that many of their primary APIs are literally screen-scraping terminal sessions and returning the screen as XML.

"I can absolutely guarantee you that many of their primary APIs are literally screen-scraping terminal sessions and returning the screen as XML"

True, but TPF is scalable enough that it doesn't matter. The API is responsive, flexible, and handles the traffic. The ugliness isn't limiting new functionality, etc.

We spend very little time talking about the cruft at the bottom. The business complexity is more important.

Things like fare classes, discount codes, availability, fare basis codes, reaccommodation, and so forth. If it were all written in rust or Go + microservices it wouldn't be any more straightforward.

Your product seems to confirm that. You aren't reducing that screen scraping any. But you're adding value.

Well said :)

> The current state of duopolistic airlines technology suppliers means that there may not be sufficient motiviation to evolve the capabilities of the old systems.

So then what is the motivation to change this now? This seems like simple economics to me.

Is the legacy only technical? I was speaking of something far broader than just the database they're using at the heart of something.

Our theories of place and travel are slowly changing due to new technical capability. Some of the lag is due to old theories embedded in old technology. But I think a lot of it is embedded in habit, custom, protocol, and business model. And in plain old not knowing. I expect we'll thrash around for decades before a new consensus emerges.

You're correct. There is legacy thinking at airlines and travel companies. It is more of a barrier than any tech debt.

I spend more time fighting "this is how we've always done it" than I spend fighting legacy systems.

>"Most airlines have pushed the legacy TPF layer into a sort of nosql data store."

Sorry what is TPF here?

The legacy IBM mainframe OS that powers the base layer for most airlines, and also things like credit card transactions.

It's completely different from the most prevalent mainframe OS (zOs/MVS).

https://en.m.wikipedia.org/wiki/Transaction_Processing_Facil...

Think of it like nosql. High transaction rates, and high contention. Great when used for it's strengths, terrible otherwise. Close to bare metal, and hard to match for transactions/second with decades newer tech. The record sizes are closely matched to hardware I/O offload cache sizes. It's an I/O beast.

Imagine an OS built specifically to ignore things like private memory space in favor of specific dedicated drivers with intimate knowledge of specific disk and cache controllers to read and write as fast as possible, skipping the CPU whenever possible. And hardware -aware record locking semantics.

Oh interesting. Thanks for the link. I guess I always just assumed it was some old IBM 360 series. Cheers.

> traditionally, seats can't be resold effectively once purchased

That's not tradition, that's self imposed to make yield management work. If you could just buy a ticket at its cheapest and sell whenever you want then yield management would collapse.

>"It may be worth mentioning that airlines are perhaps less-evil than is apparent from the outside. Most of their baffling behaviour is (IMHO) largely due to lack of capabilities rather than malevolence."

Just to recap the "Dr Dao United Airlines" incident mentioned in the article - United airlines had a passenger forcibly removed from a flight for not agreeing to give up his seat for off-duty United employees.[1]

The reason the doctor wasn't interested in giving up his seat was because he had patients to see. I think that meets most people's criteria of "malevolence."

https://www.economist.com/blogs/gulliver/2018/04/david-dao-a...

>"Airlines are very much currently burdened with a legacy that was cemented in sometime in the 50's."

Which legacy exactly are they "burdened" with? The airlines in US were deregulated in 1980s. The airlines look nothing like they did in the 1950s. The regulatory regime is gone, there's "power by the hour", fuel futures etc.

>"Ultimately, they're given a large multi-dimensional optimisation problem ..."

Yes and the airlines have pursued a policy of rapid consolidation which means forcing more people through regional hubs which are in no way optimal for their itineraries.

>"We are taking the (very small) initial steps in alleviating some of the issues by introducing increased liquidity into the market ..."

That is a bunch of double speak. Fixing the problem of overbooking is done by airlines practicing some basic discipline - "stop selling what you don't have." If you are selling seats on an A320 with 180 seats then don't sell 200 seats. In what other industry is selling something you don't to sell done as a matter of course?

The problem of overbooking is fixed by not actually overbooking. You are simply enabling airlines to continue this unethical practice of overbooking in a manner that's more profitable to them. I fail to see the innovation.

These days, it's not the legacy. It is mostly regulation pushed by Chuck Schumer[1].

Predictably, the policy has made it much more profitable to outright cancel flights at the first sign of trouble. Add to that the consolidation that occurred since then, we are left with:

> Four carriers with over 80% of domestic capacity:

> - Three large, hub-oriented, global legacy carriers (American, United, Delta) > - One large, point-to-point oriented ‘low cost’ carrier (Southwest)

> Six much smaller carriers, each with less than 5% of the market

> - Three smaller, primarily hub-oriented carriers (Alaska, jetBlue, Hawaiian) > - Three much smaller point-to-point travel merchandisers, heavily reliant on ancillary fees, so-called ‘ultra low cost carriers’ or ‘ULCCs’ (Spirit, Frontier, Allegiant)

The main legacy are the huge health care and retirement obligations. Baumol (1982) discussed the airline industry as an example of a contestable market[3]. Free entry & exit no longer applies.

[1]: https://www.nytimes.com/2009/12/22/business/22passengers.htm... [2]: https://www.eyefortravel.com/revenue-and-data-management/us-... [3]: https://econpapers.repec.org/article/aeaaecrev/v_3a72_3ay_3a...

The US airlines industry is a public utility that attempts to operate (poorly) as private corporations.

Not quite because adding new companies and routes is significantly easier than laying new utilities.

It barely ever turns a profit over the long term was my point.

>"These days, it's not the legacy. It is mostly regulation pushed by Chuck Schume"

The OPs comment was that the airlines were burdened by legacy from the 1950s. Mr Schumer's legislation is quite recent. Additionally the legislation "Passenger Bill of Rights" is quite reasonable - let your customers off the plane after 3 hours of delay. It's pretty easy to avoid that penalty. It's also worth pointing out that legislation was in response to unreasonable practices by the airlines in the first place. I'm not sure why it is even relevant here.

>"The main legacy are the huge health care and retirement obligations" How is this relevant to discussion about the practice of intentionally overbooking?

So-called "Passenger Bill of Rights" takes away options from passengers[1].

    "We found that for every minute of tarmac time being
    saved there is, on average, a three-minute increase in
    the total passenger delay," said Vikrant Vaze, an
    assistant professor at Dartmouth's Thayer School of
    Engineering who co-authored the study.

    Adopted in 2010, the Transportation Department rule
    requires that passengers be allowed to deplane within
    three hours of boarding or landing for domestic
    flights. Airlines are subject to fines of up to $27,500
    per passenger for violations.

    Hoping to avoid fines, airlines have been quicker to
    pull the plug on flights, Vaze said. That forces
    passengers to scramble to find another flight to reach
    their destination, something that has become
    increasingly difficult as industry consolidation
    reduces the number of available seats on planes.
    Flights that are not canceled have to start lining up
    anew after letting passengers deplane and then reboard,
    adding to the ultimate delay.

    "Most of these extra delays are being felt by those
    exact same passengers," Vaze said. "It's just that
    they're not on the tarmac."

[1]: http://www.chicagotribune.com/business/ct-tarmac-delays-0105...

Again there is no reason the airlines can't let people off the plane after a couple of hours in order to avoid penalties. And it's unclear why you are holding this up as "burdensome" legislation.

Just FYI, quoting like this makes it impossible to read on mobile. Better to wrap in asterisks.

I'm not sure that "cancel[ling] flights at the first sign of trouble" is the issue either.

I had my flight on Air New Zealand cancelled after four hours in the air, forcing us to turn back due to a mechanical problem. After three hours on the ground, they gave up. I ended up with a night's stay in Auckland, a free breakfast, a new flight to LA first thing in the morning and double points. I'll probably be an Air NZ customer for life as a result.

Even United can sometimes rise to the occasion. They cancelled our flight to San Francisco because of SFO's typical shenanigans with fog and not allowing small commuter jets in. We got rebooked the next day, though all they had were economy class seats. We said fine, at least we're on our way. We got upgraded at the gate and we didn't even ask for it.

If a company wants to make customer service a top priority, they can do it.

>You are simply enabling airlines to continue this unethical practice

Don't project your morals onto the rest of us. As a frequent flyer who likes to travel I love this practice. It means lower prices overall and it gives me the opportunity to fly for free somewhere else later if I volunteer out. With the switch to real cash compensation on some airlines, I can even make money if I don't feel like using it to fly.

I have flown over 2 million miles in my lifetime and I have never seen someone forced to give up their seat. The airlines always enticed people to volunteer out with (sometimes significant) compensation.

United behaving like assholes with their forced removal is indicative of their behavior, not the practice of overbooking.

I'm fine with increasing required payouts for forced removals to deal with piss poor airlines that don't conduct an auction until volunteers are found, but banning the practice outright is throwing the baby out with the bathwater due to ignorance of the practice.

>"Don't project your morals onto the rest of us"

I'm not projecting "my" morals onto anybody. Intentionally overbooking and knowing full well that you are going to bring inconvenience and stress to people is just plain fucked up. That's not really a grey area. If you bought a ticket to a movie and then were asked to leave the theater before the movie started in order to make room for someone else who bought a ticket you would be livid. The same is true for any other "ticketed" event.

>"I have flown over 2 million miles in my lifetime and I have never seen someone forced to give up their seat."

Here you can read all about it:

https://www.cnn.com/2017/04/10/travel/passenger-removed-unit...

And that is not the first time either. Just because you haven't seen it doesn't mean it doesn't happen.

If a significant fraction of people won’t show up AND a significant fraction of those who do show up will have flexible plans and be happy to take a later flight, “overbooking” seems a lot more reasonable.

>"If a significant fraction of people won’t show up ."

I think its safe to say that there isn't a "significant fraction" of people who don't show up for their flights.

The fraction is obviously significant to the people trying to make sure airplanes are fully utilized, so this seems like a word game to me. People miss flights all the time.

A brief look for stats doesn't turn up hard data, but TechCrunch has it at ~5%:

https://techcrunch.com/2017/04/11/overbooking/

Someone who worked in ticketing for two airlines says it's 2% for flights to hubs, and 5-8% for flights from hubs, because of missed connections:

https://www.quora.com/What-percentage-of-people-who-check-in...

According to IATA, a good year for airline industry profits is circa 5%:

http://www.iata.org/pressroom/pr/Pages/2016-12-08-01.aspx

So even if we take the lowest fraction, 2%, filling those seats represents 40% of the profit margin. That sounds pretty significant to me.

I think it's safe to say you don't understand why the refundable fare class exists then.

There's a price for which I'd be delighted to be kicked out of a movie or off a flight.

For a movie, it might be only $50. For a flight, it could range from $1000 on up, but it's a pretty rare flight that I wouldn't give up my seat for $15,000 and rebooking. I suspect it's even more rare that no one on a flight would take $15K plus rebook.

Airlines would be willing to do this if all tickets were 100% non-refundable, non-exchangeable, no excuses. But, that would be a terrible experience for customers. Imagine if you missed your flight, and were informed that you had no choice but to pay for another ticket.

Movie theatres can do this because the $12 loss of missing a movie is annoying, but not life-changing, and the social contract is that event tickets are not changeable.

I would wager that the fallout of tens of thousands of people per day being told they have to re-buy a ticket if they miss their flight would have much greater negative fall-out than the much smaller number who get involuntarily bumped.

>"Airlines would be willing to do this if all tickets were 100% non-refundable, non-exchangeable, no excuses. But, that would be a terrible experience for customers. Imagine if you missed your flight, and were informed that you had no choice but to pay for another ticket"

In the US all three big carriers - American, United and Delta all charge $200 to change non-refundable tickets.[1]. Any you pay the fare difference if any on top of that. So there's already a hedge in place.

[1] https://www.forbes.com/sites/alexandratalty/2015/07/20/dont-...

Yes, but they don’t usually charge you this fee if, e.g. you miss a flight.

Additionally, that $200 fee comes nowhere near covering the lost revenue if you change flights a day in advance of departure and that seat isn’t re-sold.

I hope you can see that a last-minute change (even with a nominal fee) means the plane takes off without a paying butt in that seat.

If you miss your flight due to a delay by the airline such as missed connection then you will not pay this penalty. If you miss the flight because you overslept or got stuck in traffic then yes absolutely will pay this penalty. You will have to pay this fee at the counter in order to be rebook. I also wouldn't call $200 plus any change in fare for the rebooked flight "nominal."

No, those are your morals. And you're projecting them on to everybody. "Just plain fucked up" is another way of expressing your personal values, not a rational argument.

I get that you value the "but we had a deal" criterion for fairness above the other things people value in this situation. Which is fine! You do you. And I agree that forcible removal is a bad idea that we should ban.

But booking so that the airplane will probably be full and paying people to wait a bit when statistics make it overfull is something that the great majority of the public is fine with. Flexible people like it quite a bit.

>"No, those are your morals."

No its common decency and etiquette. These aren't "mine" they are social norms.

>"I get that you value the "but we had a deal" criterion for fairness above the other things people value in this situation"

Sure everybody is "fine" with this as long as its just chatter on HN. It's all optimization problems and capturing true market value etc. But I'm willing to bet that when you yourself are on the receiving end of an involuntary bump by an Airline your first thought is most certainly not "I'm fine with this."

> Intentionally overbooking and knowing full well that you are going to bring inconvenience and stress to people is just plain fucked up. That's not really a grey area.

First, passengers who buy refundable seats frequently cancel last minute to take a later flight. The airline can somewhat estimate how many cancellations any given flight will have and overbook by some amount less than that. This happens all of the time (nearly every close to full flight you are on will have had this happen) and you don't even know about it. You just benefit from it in the reduced fare due to the increased margin this provides.

Second, even if more people show up than projected, this doesn't bring "inconvenience or stress" to anyone in the vast majority of cases because they begin an auction to give up seats. The people who volunteer out come out ahead winning an auction they willfully participated in by selecting their own price. Delta will bid up to $10,000[1] to get people to release their seats. That's more than a month's salary for most people in the US just to take a later flight. Nobody feels like they got screwed in the volunteer situation, because they volunteered in exchange for the money/flight voucher.

>Here you can read all about it:

Cool, you sent an article echoing what I already said about bad actors like United's regional carrier who don't define the norm. That case is not what happens in the vast majority of overbooking situations.

You ignored what I said to make what point? You are intentionally confusing forced removal with the standard case for overbooking where people volunteer out and come ahead.

>I'm not projecting "my" morals onto anybody.

These are your moral issues which seem to be driven by ignorance of how overbooking works in 99%+ of the cases. I suspect you have flown very little or have never paid attention at the gate. Please don't armchair referee an industry practice you don't understand that benefits both consumers and the airlines in all but pathological cases.

1. http://money.cnn.com/2017/04/14/news/companies/delta-10000-o...

Reply on news.ycombinator.com