> I think all important things in travel have been invented.

In decades, if not years, you will look back and laugh at yourself for this.

Let's see:

- When should I buy a ticket, what is the punishment for not buying right now?

Kayak has a "when to buy" feature, but I can't think of anyone else. Microsoft Travel stopped doing it and even then there's was only for US.

- How often is the flight I'm booking delayed? For how long? Airport or Airline fault?

Some Airlines show you the delay/arrival percentage but its not aggregated anywhere AFAICT. Also I believe airplanes have a lot of leeway before a flight is considered "delayed" Also people suggest never to land/fly out of O'Hare in Chicago if possible but is there evidence for this? Would someone outside the US who is not a frequent flyer even know about this?

- How much space does this airline's seat provide?

I'm willing to spend a little more for more space on longer flights...

- Does this plane have power outlets? And not the ones that EmPower plugs, those don't count. I'm looking at you United!

The problem is there's actually very little accurate information except the cost of the flight, and thats what most people choose to pick their flight. Sometimes even willing to take weird layovers/extra stops to do it. If it can be shown that your comfort and stress would be better cared for by purchasing a ticket that cost $50 by next Wednesday.... well no one has done that.

How about this:

If you travel quite a bit (say 10 international flights/yr), would you pay 15k/yr to never have to think about travel? You book meetings on your phone "Jan 20, Calcuta | Feb 23, Tucson, etc | OScon in X, etc" and everything else happens behind the curtains.

The startup books flights, hotels, taxi, shuttles, trains, etc, and your Google Now just talks to you: You have a flight tomorrow at 8:00. When you land: Take train A293 at 3pm, etc.

The more you travel, and based on feedback, the app learns (less train, a little more up-scale hotels, she doesn't mind taking the bus, etc)

Like being a millionaire with a full-time assistant, but for the middle class. Does that exist? Could AI tackle it? Maybe a semi-automated version?

Unfortunately, established players often means full of legislation protecting those players. There are plenty of ideas in the travel space - but they require the ambition to challenge governments and lobbying. Uber, Lyft, and AirBnB are a great examples.

> what is left to invent in this area to be so disruptive so that a start-up in this area makes sense? I'd love to be contradicted, but I think all important things in travel have been invented.

Not even close. Travel is sooooo god damn complicated and inaccessible to most people. Even though traveling should be popular, on a population basis basically nobody travels. This can change with better software, I'm sure of it.

You're travelling to a new city you've never been to before and you're looking for a hotel that's cheap but decent quality, in a safe area and well connected via public transport to the tourist areas.

How many steps does it take you to find a hotel ?

How many steps should it take ?

I can think of two very easily: Sub-Orbital travel and medical tourism.