Greetings Ft'ers,

I'm curious... If you could start your own airfare booking agency, what new features would you include? How would you make it better than any other website? Can you think of something innovative that would make it really unique? What would be something incredibly useful that isn't offered already on other airfare websites?

In general, how can you make a better airfare website?

The things that would make a site great for me would require queries that are too expensive, both from a GDS cost and a processing cost. Things like DreamMaps that also query availability instead of just published fares. More often than not I take a trip because the price is right, not because I need to be in some city. If I want to have a long weekend (2-4 nights) from one of 3 departure cities to anywhere the processing required to run that query is pretty high. But that is the search I'd like to run.

I pick my routing, I pick my carrier for each flight and I pick my class of service for each leg of the journey.

At it stands now, I can do a little bit of choosing, but not all of it easily.

I completely agree. Most of my travel is for work but I'm always looking for inexpensive flights to somewhere for a weekend. I know it's unrealistic but I would love to have the ability to put in dates and origin and the website would search either all cities serviced by an airline or a handful of cities that I put in. Again, I know it's unrealistic but you asked.

Yup. Same here - I want to see where I can fly cheaply. More importantly, however, is that it is not just domestic - but international as well.

If there was a way you could then convert that into some kind of cost per mile figure, you would make me very, very happy!

To be honest, I was happiest with the old EaasySabre interface where you had the flights and the seats available in each bucket, all available in a nice easy command line. Not too complicated a whole search system - just enough to let me find what I want quickly.

It would also be nice to have not just seat charts, but a list like SeatGuru which gives you an idea how much room you get with each class of seating.

"One-to-many" routing, especially internationally; for instance, show all fares from a given North American airport to any Western European gateway. (Some sites sort of do this, but I have yet to find one that does it fully and easily.)

"Open jaws" fare quotes. Put in for instance an itinerary like outbound ORD-LHR return PAR-ORD and get fares for it, that is, quotes for round trip fares that allow open jaws without costing much more. Many airline sites do this, but not most comprehensive sites, and some that do just seem to give you two one way fares.

Showing for any multi-segment itinerary in any fare class if stopovers are allowed and if so how much it costs extra.

Quote special two-for-one companion fares offered on any itinerary in any fare class.

For any quoted fare allow an easy way of answering the question, "Is this exact same ticket also sold by another code-sharing airline and if so what is the cost?"

In theory, FareCompare does this, but the pricing is unreliable.

Most of what you're describing can be done with ITA, but it is not a booking engine.

Never mind - remove post.

Quite the question, since I'd love to have a site that automates what I otherwise will spend hours doing manually:

- Start with a departure city or range of departure cities.

- The ability to pick a destination and have the site run a recursive search from the departure airport(s) to every possible destination, recursing down to repeat the search for those airports, returning after a preset number of max connections on potentially (but not necessarily) separate tickets to list all possible routings from departure to destination with associated total cost (give or take baggage charges, etc) listed nearby.

For example: I wanted to book a flight from ORK to POZ, but balked at the �250 cost for a direct one-way. After some looking, I realized I could route ORK-MUC-POZ on a combination of airlines to get a one-way for �100. For a few more connections, a RT via DUB, ZRH, and TXL would have come out to somewhere near what the direct OW would have cost.

- The option to link nearby airports by non-flight transport would also be nice. Just in case a large savings can be made by doing a transit from, for example, TXL to POZ by train rather than flying directly.

- The ability to input an origin and budget and return a list of possible destinations, filtered by carrier/alliance of preference.

For example: I have X days off and �300 to spend on flights. Where can I go?

- The ability to only return a confirmation number would also be great, but that's only for people who like a surprise when they check in.
"So you're going to Athens today"
"I don't know. Apparently!"

- The ability to white-/black-list carriers. I want to find a trip to X, but I'll walk before I fly FR, for example.

- The ability to specify booking classes. I love how many sites will automatically break it down to show cheapest, flexible, etc. But I want one seat in discount Y and one in slightly-more-than-discount Y, as I need status miles but my co-traveller needs the cheapest fare.

- A column listing the mileage accrued on your FFP(s) of choice per fare class. Ideally one listing the total mileage for each routing as well.

- Although it may only be applicable for a few, the ability to book an itinerary but restrict it to as close to a predefined itinerary as possible. Alternately, to specify two itineraries and force the booking engine to keep them as identical as possible.

For example, I book one itinerary for my girlfriend to go to a conference, with me meeting up at a later date and the two of us flying back together, potentially to different destinations. Things like forcing overnights where one overnight might be forced already, etc. We do this every now and then, and it's just that added bit of effort to try and coordinate as much of the journey as possible.

- A "minimum miles" routing tool would also be helpful. For when you need a MR and just specify "show every routing that is more than x thousand miles".

I'm sure I'll come up with something more, but there's lots of web tools I'd love to see for this!

This is a great idea, but to make it even more valuable, I'd want it to specify not just accrued mileage, but also elite-qualifying miles. It would probably take quite a bit of programming to get there, but with the increasing prevalence of fares that earn RDMs but not EQMs (or reduced EQMs), this feature would be huge.

Oh, definitely! And minimum-miles rules on top of that, some of which will vary based on your FFP and booking class.

- As others have said, the return of Travelocity's "Dream Map," but worldwide, and one that actually works and actually has seats at the fare quoted.

(VERY challenging due to the computational load required to analyze every destination from San Francisco to Sacramento to Santa Fe, determine the fares and then present them all.)

- Some mechanism to determine what airlines service what destinations on what days out of which hubs. Recently I was trying to determine who serviced the Cayman Islands. Once I figured that out I needed to figure out what hubs they flew out of, and then on what days they departed from those hubs.

This kind of search could be done in advance and cached, with hourly updates. Sure, you'd make an occasional error by presenting something that's no longer available (or missing something that is available), but availability can be checked when the user selects the flight.

How? The amount of data to cache hourly would be geographically staggering.

Let's say I want to fly out of PDX, and the dream map options are $250, $50, $750, $1000 and $1500. Every hour the system needs to calculate every destination from PDX, from San Diego to Reykjavik to Salem to Naples (both of them) and cache the fares, then do the same thing for every other airport code.

It's daunting - I work in the software field and I'm 100% confident this is why Travelocity dropped the feature.

Parallel Processing?

If you break it down to setting each core to calculate one destination and sub-destinations from there, you can reduce the processing time by an factor of the number of cores working on the problem.

Spread that work over a number of X CPUs with Y cores and assuming each core can run query as fast as it would take for a single system to run the process, you have an X*Y speed increase per City.

If you're willing to trade absolute accuracy for speed, you could take each routing as you calculate it and compare it to the inventory for that segment. Then based on the inventory, it should give you an idea as to whether or not you need to update the cached data the next time. Y9 on all legs probably means you're not going to need to check again in the next hour. If someone picks that destination, then check it again.

But I've avoided doing anything serious with software in a while. I just know that threaded, multi-cpu systems are where you want to start if you want complex calculations done fast.