I'd be very interested in knowing what your site was. My email can be retrieved via my profile/blog. Maybe my similar experience can help you.
In a previous life, I ran tag-board.com, which placed a little javascript widget on your website that let your users interact with each other much like an irc channel. This was before ajax was even coined a term. The site still exists, but it's run down and I don't have anything to do with it anymore.
When I first started it, I never thought more than 100 people would ever use it. At it's peak, there were 600k accounts and tons of traffic that went along with that. At first I didn't care to make money, I was just happy that people used something I created, but when I realized I would need more than a $5/month hosting account, I started looking into options.
The first plan was actually a premium model - hack up a bunch of new features that would only be available to paying users and sell accounts for $20/year. It worked, and might for you as well. At some point, I had over 1,000 paying users.
After awhile, the problem was that hosting costs of the free accounts were eating a good part of the profits on the paid accounts AND I was spending a ton of time on maintainance. I felt responsible for pretty decent uptime/performance/bug-freeness since I had people paying money. A friend of mine pointed out that the amount of time/stress I was putting into this wasn't worth the money. This made a ton of sense to me. At that point, I strongly considered just keeping paying users alive until their year ran out, and then axing the whole thing. I then realized that if I was willing to get rid of all my free users anyway, I might as well try something aggressive - I found an ad network and ran popups on the non-premium accounts. This was ~2002. To my surprise, I got fairly minimal grief, more premium signups, no noticeable drop in usage, and enough income to make it worth my while to keep going. This lasted awhile, but popup revenue slowly dried up - firefox and other browsers started killing the popups. I had handed the whole thing off to someone else well before that happened though.
If I were to do it again, I think I'd do the obvious thing that nobody recommended and just charge everyone for the service, although a much smaller fee (maybe $5/yr). I'd let people sign up for free, use the account for a month, and then switch it over to showing an "ad" of sorts on the owner's website where the widget otherwise would be. Clicking through would let the user interact with the widget, but on the "ad" would also be an option for a user of the website to buy the site owner a subscription to this widget. This would dramatically increase the number of potential customers and would essentially be like a tip jar for the website owner. Users could buy the widget as a gift for a website they like to use. At the same time, I wouldn't try to stop people from just creating a new account and getting another month of free usage, but I'd annoy them forcing them to do this every month.
I don't actually know if this would work, but I think it would. I suspect there would be fewer users. However, as a side benefit, you would know what this is worth to those users, which is a nice feeling. The negative here that scares many people from going this route is that if you suddenly start charging, you really do have to provide a service. You can't go down for 2 weeks. And you have to do a little bit of customer service (emails, refunds, fix bugs) or find someone to do it for you.
One kind of ad you could run that wouldn't have trouble with cramming into a widget is popups.
Don't downvote me because you hate popups, so do I. It's a legitimate possibility that you might consider looking into. If it's a question of shutting down the service because of bleeding costs or running popups on 5% of pageviews, its an interesting question.
I've long wondered if this kind of approach is actually something that affiliate programs would allow if you got big. You are not promoting their wares, you are simply profiting off of organic promotion that would have occurred anyway. I wouldn't be surprised if Amazon or whomever decided to close your account if you started making significant amounts of money.
Maybe a minor quibble, but isn't DDG powered by Yahoo BOSS with features added on top? Yet, Rob then goes on to say "Yahoo can't be taken seriously" and points out queries where Yahoo does a poor job of handing synonyms - DDG has the same problem for the exact same query if you try it. Furthermore, the other queries which he suggests give irrelevant results on Google seem to give me irrelevant results on DDG.
In a sense, Google already has something like this, the Custom Search Engine: http://www.google.com/cse/ - the interface is a little different though, you create a "search engine" which has something like the slashtags built in, rather than specifying a slashtag as part of the query. Has it's pros and cons.
I'd prefer to see these where each step in the algorithm was a fixed width that is the same across all of the images. As-is the step width is dependent on the number of steps. This fails to show the speed of convergence.
But watts/dollar is quickly decreasing. I wonder if performance per dollar spent on energy is increasing or decreasing. I wouldn't be surprised if it's decreasing.
I think they are trying to point out the options - MS office vs. Open Office. People visually scan lists like these and ignore the rows where the both products are identical. If they checked that MS supported open office, the natural way to read this list is that Ubuntu doesn't support office of any kind, which isn't the point they are trying to make.
Not likely since the j/s is hard-coded into website templates. Even if they did change the domain, very few websites would have the new domain for awhile.
Not exactly correct. If the URL you click on is HTTPS, then the referrer is sent. HTTPS->HTTP no referrer. HTTPS->HTTPS referrer (even if the domain changes). The encryption is preventing the network from seeing the data, not the remote hosts. If the browser sends the referrer encrypted, even to a different remote host, the network still can't see it.