IMO, the best way to make travel "social" is to make social travel easier, viz., make organizing trips for groups easier. That's a legitimate problem to solve.
I don't think any app is going to effectively take choice and discussion out of the equation. Attempting to do that, especially with a category like leisure travel, is asking people to let an algorithm plan their vacations. Most people won't go for that; they want a large degree of choice and consideration in the process. But the stuff nobody likes doing -- planning flights, searching for fares, booking rooms, etc. -- is where the pain points reside.
Regardless, there are a lot of people who may be going to a suburb and not a hotel by the BART stop that don't want to do 3 transfers and would rather take a shuttle. Until a multimodal search engine can show me those options, I'm not really interested.
However, this was often the premise of these inspiration sites that didn't work out. Wanderfly has the ability to search by number of days and how much money you had to spend, but it didn't really work.
These things are good in concept, but it gets back to my point, that they need TONS of data to make any meaningful recommendations, and that data doesn't exist yet on travel. In many cases music is easier, as a former classical musician I can attest to this, in music theory there are diminished chords, augmented triads, all different types of existing classifications of different types of sound.
In travel it's a big mess. You need big data solutions to sort through the problems, and just building a pretty interface won't solve it, which is what most of these companies have been doing.
- Skyscanner, as David says - Adioso (YC W09) - TripCommon - you can also set up alerts 'SF to Europe for less than $Y00' - Getgoing (YC S12) - Kayak explore
I've always found Skyscanner has the best results, but I'm curious to see how newcomer Getgoing compares.
Google Flights has something similar, it lets you see prices for destinations:
Plenty of room for innovation in this space.
To contrast this with music, where a song will always stay the same and the 3 minutes of it are, arguably, easy to deconstruct into components, classify by genre/rhythm/artist/etc., and then index.
Disclaimer: we too are building a "trip planning tool" but I think we've found a good angle to approach the problem with to avoid the situation most "inspiration"/"social"/"planning" travel startups find themselves in.