I played with this idea when I was doing an algorithmic composition project. I started with a Markov chain analysis of a simple Chopin etude, which gave me parameters, such as:
- The distance between the lowest and highest note is 2 octaves
- The key won't change, so there are only 8 notes per octave and given the first parameter, I only had 17 notes to choose from.
- The shortest note was an 1/8th and the longest was a half. Most were quarters and there were a few dotted quarters (if I recall) so there were 4 note lengths to choose from.
- The main melody in the original was 4 measures (again, if I recall) so my generated melodies were set to be that length, and always end on the root
Given all of these parameters, there were still many thousands of combinations. Applying the Markov chain limited that a lot, but that was only to make it sound like Chopin (which it did, a little). However, most of the thousands of combinations sound like crap, which is not an insignificant point. If you're willing to call some random sequence of notes a melody, then there are an infinite number
Focus on standing up. Alarm clock is set to the news and I'll listen for anything interesting. Use toilet, drink glass of water. Go to kitchen and get breakfast going for family. Only then do I check email/messages for anything urgent. I used to be more vigilant about checking email first thing in the AM but not now.
My most reliable clients have been through contacts I made when I had crap cubicle jobs. I don't know your situation but it may be worth snagging a cube job long enough to meet people. And it may be totally counter-intuitive and potentially mind-crushing, but the people in sales meet all the clients who have money to spend on development. If you could find your way into the social events that sales people go to you'll be halfway there. It's cynical, but getting gigs depends on people seeing you as "a good guy" more than whatever skills you have.
That said, one trick I learned was to always scan the local news looking for local businesses who recently got grants or some other windfall of funding. I assumed (usually correctly) that they would spend some of that windfall on building or revamping their website. I got some good long-term work that way. It was a cold call, but one with research behind it
This parallels my own experience. I too had a "can-do" attitude 10 years ago but now just can't be bothered with many things that I used to spend time on. This, however, is not laziness, it's maturity. You have now reached the level of maturation in your understanding of technology to know that things you learn for a one-off project will be forgotten within 6 months. And you are nearly at the level of maturity in relationships to know that if your girlfriend is truly passionate about selling her jewelry that she can learn to maintain her own eStorefront
This is hardly a new idea. You could argue that many of Dickens's novels were written this way, bit-by-bit through installments in the newspaper. The format led to the cliffhanger nature of his stories.
Look at NaNoWriMo or DailyLit for current examples
In brief, I would look to your community, whatever that is. Ideally you live in a place with an active development/startup culture with lots of talented and motivated people, and all you have to do is lean out your window and ask technical questions. But it's not 1998 anymore.
Do you have an alumni group or some kind of online social network you rely on? When something big is at stake, it's risky to trust a complete stranger. You may both have the best of intentions, but they could end up being not having the skills or (maybe more importantly) the personality you need.
Craigslist, from my experience, leads to more misses than hits. LinkedIn is just beginning to be a non-worthless social network.
If you've put the feelers out and no one has emerged, one last resort could be to get some college interns. Some colleges will give credit to students as long as you fill out the paperwork to call your job an internship. I did this in New York City long ago. We had to pay minimum wage, but in return got an intelligent, hard-working person who (for better or worse) was not vested in the project long-term
Given all of these parameters, there were still many thousands of combinations. Applying the Markov chain limited that a lot, but that was only to make it sound like Chopin (which it did, a little). However, most of the thousands of combinations sound like crap, which is not an insignificant point. If you're willing to call some random sequence of notes a melody, then there are an infinite number