4 are folders for random hacks or experiments I do (eg, one for Ruby, one for Clojure, two for misc stuff). So I won't count them.
5 are random admin stuff, so I won't count them either.
2 are folders containing work from previous jobs. 2 are university projects.
That leaves 20 personal projects I started.
Unshipped/unfinished:
-Botnet (Facebook game in Rails, never finished)
-KAF (Business plan + notes for a startup in the smart TV space. Never actually wrote any code for this so I don't know if it counts as a project).
-NAND cpu (simulated CPU built from simulated NAND gates. I got as far as making an adder and got bored)
-Pixelworld: (a multi-user canvas thing.)
-Robots Rising (clojure browser game, got bored and gave up)
-Unimemes (can scrape Facebook groups and find the most liked pictures, got bored)
"Shipped" but gave up quickly:
-ChineseVille (Facebook app similar to Lingt, "launched" with a very incomplete alpha and gave up after no-one showed interest)
-Discrete (Pixel-art webcomic, did about 5 strips, some people on Reddit liked it but I decided it was too pretentious so deleted it)
-Nohao (user-rated learning resources. Released a very early version and no-one was interested, so gave up).
-Real Reading List (similar to Nohao, but focused on textbooks. (lesson: everyone has had the idea of reselling textbooks to students, it has never worked)).
Learning projects:
-Facebook-friends scraper
-Pond (Clojure game)
-Rails survey app
Hackathon projects:
-Twitballot (twitter polls)
-Vettr (online careers fair)
Shipped:
-My personal homepage
-Shintolin (MMO browser game)
-Writing (stuff on my personal blog)
-Warwick Fashion Group (blog done for a friend)
-Warwick Uni Startups (website for a university society)
I count the last 3 groups as successes, which makes 10/20 hit rate. That's better than I thought, though I'm being a bit generous with the definition of "shipped". Shintolin is the only significant personal project I've worked on (it was also the first, go figure). There's also a whole bunch of work/university/freelance projects I've completed which aren't on this list, but that's obviously thanks to massive extrinsic motivations.
Unlike most people, my problem is shipping too early -- I tend to release stuff in a really crappy, unfinished state, when of course no-one will be interested. (It did work for Shintolin, though -- I posted it on a forum when all players could do was walk around and punch each other, and people seemed to really like it). That said, I don't regret giving up on any of the other projects (except maybe ChineseVille, which could have worked if I stuck with it).
(And of course, software projects are by default never finished. That's why I've come to enjoy writing -- you realise you've already written too many words, even though you have more to say, and can then stop and hit "publish". There's no limit to the features you can add to any piece of software).
I've realised I'm gonna have to partner with a finisher if I want to achieve anything significant in future.
Here's a few that come to mind:
'07 - A Search Engine that used MS/GOOGLE/YAHOO as my seed and resorted the order based on some composite ranking from comparing the three results.
'08 - An Email Myself Blogging service that took some words, links, and posted it on the date in the email.
'09 - A Local Coupon site to work against Valpak. You could throttle the amount discounted so you didn't have the Group On effect. I came back to this idea 20x, but in order for it to work you have to hit the pavement and talk to store/rest/bar owners.
'10 - Social networking site for fitness with a workout tracking tool.
'11 - Quit a company that got bought by Google (just had to get it off my chest)
'12 - Notes sharing tool that's transformed into something else. I stopped for a month on it, bugged me so much that finally late March I got back on that damn horse. In 4 weekends I have POC (like a lawn motor on wheels) that finally worked...but I'm excited about it now.
In the end that # of projects doesn't count for a damn thing. Everyone of those ideas listed above could work, have worked for someone else. Someone that saw one thing to the end. You aren't alone though, so don't beat yourself up one bit.
1, that's the number I am going to focus on this year. Finishing one idea out.
Eventually you learn: If you build it they will come; If you are lucky.
Burnout is a real killer
In my case, for the most part, I'll either come up with an idea or a challenge. In some cases, I'll work towards an idea and usually, I'll get further along with those because I start with the boring stuff like logins, user profile pages, etc.
With the challenges ("How would you build a web site that did <insert something novel here>?"), those are pretty much guaranteed to be DOA. I do the hard part, sometimes realizing my initial approach was stupid, starting over, and then doing it more elegantly, and then uhhhh, well, that's it. The chance of me writing all the boring crud for something like that to make it multi-user is infinitesimal.
