I do think that they are shrinking the market. One of the latest Planet Money stories shows how much that market has shrunk: http://www.npr.org/blogs/money/2012/01/20/1454660...

I think that's why you have to concentrate on things that the big competitors can't copy or react to easily, which we've tried to do on DuckDuckGo. Generally that's not for technical reasons, but for business, legal or cultural reasons.

Privacy is a good example. I'd argue that even if they go that far to react (which I think they won't for various reasons), it still would't be enough to appeal to all the people who really care about privacy. Those people generally either a) would like to use a company that makes privacy a core value and/or b) like to use a company that is solely focused on search.

Np -- I'm adding to investigation list.

Right -- well I try to start really simple, and that's what we have now. Everything is modular so that things can run on separate machines, but most run on just big instances. We use round robin dns with short ttls and have failover records for the IPs so if something goes down it just goes to a mirror. Not sure if you have specific questions :).

We use Asana, and we like it. It's free so I didn't include it, but maybe I should have.

Wow, that's expensive! Though does look cool. Part of the issue I have with these things is they don't seem super tuned for the cloud (EC2) life where new servers come up all the time and they might be relatively small (thus not warranting the huge price).

Yes. For now check out this old one for early on: http://www.gabrielweinberg.com/blog/2009/03/duck-...

I gave it a go a while ago, as well as the new Opera mail, Fastmail, Hotmail, Yahoo and Neomailbox.

I don't trust them at all, but I don't have any good alternatives in terms of achieving productivity (keyboard shortcuts, filters, ui, etc.). Any suggestions?

I thought it was a better mix of those services, traditional forums (phbb), q/a (stack exchange), and social voting (reddit). It does all of those things OK, though none as good as the specialty services themselves.

Ultimately, I was looking to get a community going, and that did work.