Qwant was launched in January 2013 having been in development for a couple of years. It's an interesting engine, because it really tries very hard to do a whole range of different things. It says of itself "QWANT offers the first web and social service, where you can dynamically use the power of your own brain to refine search in classic Web, Live, Social, Media and Shopping verticals so as to reach exactly the information and the people you are looking for, those that answer your exact query of the moment." What this means in practice is that you run a search and get 6 datasets to play with. It looks a little like this:
A series of images/video results across the top, and then 5 columns - Web, Live, Qnowledge Graph, Social and Shopping. They are very cramped, but you see enough to let you make a reasonable choice. However, if you want a different layout, these are available: classic (as above), mosaic, media, people. The results will be reformatted for you according to choice. Results can be bookmarked and saved, which is a nice function and there's also a 'hot trends' feature, which I suspect is being pulled from Twitter.
Searchers can also refine their results as well, though I didn't find this worked terribly well, and I kept getting no results when I should have got several (starting a search with 'library' and using CILIP to refine for example) and I wasn't happy that the engine reinterpreted my search without giving me the opportunity to override - turning 'CILIP' into 'clip' for example, which is irritating and unnecessary. It was also a shame that there wasn't an Advanced Search option, and they really are not clear on exactly where their results are being pulled from.
In summary however, it's a good start, and I'll be interested to see exactly how it develops. Certainly worth giving it a go and if you do - let me know what you think!