I'd suggest staying with whatever keeps it fun.

Selling your game means a lot of extra work. You'll have to support them, which means testing and the frustration of dealing with bug reports from the nearly computer-illiterate. It means spending a lot more time with people, and with already-finished games, which unless you quit your day job means you'll have less time to work on new games.

I've spent a lot of time thinking about this, even though I wouldn't even consider it until I actually *finish* a game. With things like Java Web Start, Xbox Live, and Valve's Steam it has never been more convenient to publish your game yourself. Paypal or Amazon can handle the purchasing stuff, so all that's left to you is to make the game and support it. It's a powerfully tempting idea.

The problem, as I see it, is that there's probably just as big of a gap between my newbie estimates on how much work is involved and the reality of the situation as there is in typical newbie estimates on how much work is involved in making a game and the reality of that. Sure when someone is pitching an idea and can sit down over coffee dreaming things up it all seems pretty easy, but sit down in front of a computer and try to make it happen and you quickly realize how much you've assumed or didn't think of.

But, if you're asking for a suggestion (and you are, this is the internet, everyone is entitled to my opinion if they want it or not :) I'd say stick with what keeps things fun. Maybe that means taking that extra step and really trying to make a business out of this. You'd probably have more people playing the games, and maybe seeing that happen (plus the money, money never hurts) is worth sacrificing some of your precious time on the side. Maybe it means staying with free games. Whatever floats your boat, go for it. You've obviously got the talent to make pro work happen, but you don't need ego stroking from some guy on the internet to see that.