Since building stuff is both a career and a hobby, I've always had 3 types of "projects": * One or more services that are intended to make money. Some of them

Since building stuff is both a career and a hobby, I've always had 3 types of "projects": * One or more services that are intended to make money. Some of them never make it; I've more failed ideas under my belt than winners. But the winners really win and pay for all the time spent on the rest; Improvely and W3Counter are doing really well and provide a comfortable lifestyle financially. * Projects that are never intended to make a profit. I have a few open source libraries used by thousands, but I never intended to build any business around them, and I don't ask for donations. One did once generate so much customer support e-mail and customization requests that I made a commercial version that sold well, but I was never passionate about it so I ended up selling the rights and probably wouldn't bother doing that again. * Projects that I know full well will run at a loss, but I build them anyway for various reasons, from "it's fun" to "maybe it'll be good marketing for the paid stuff". I've hosted website guestbooks for people since 1996. I run the contact form and some other widgets for about 4000 Shopify e-commerce stores, which costs a VPS and some Mandrill/Sendgrid fees to send a few thousand e-mails a day. I just felt like hacking on their API.