I've found that having a plan helps a lot. One of the problems with projects is that without a plan you have no way to know what you are interested in building. There's also no way to determine if you are making progress so you give up. A plan solves those issues. A quick five or ten minute plan will do.

The other issue is that you need to solve, "the give up problem." Everyone will eventually give up on all projects. You want to forestall that by working as fast as possible to finish your project. You are literally racing against the give up point. You want to do that by hitting goals in your project as fast as possible. I can tell you that if you are making progress you are less likely to give-up. In the past I was more leisure about my projects. I can get to it tomorrow, I used to say. The problem with that is that if you do that too much eventually you'll just get bored and give up. You don't have to kill yourself but you do have to make progress at a frisk pace.

Are you doing everything by yourself?

I do a lot of stuff alone and get it to a base level of functionality. Then I have some friends who I trust won't dis me for sport and have them ask critical questions like:

* What benefit will the user get from this thing?

* Are you the only target market for this thing?

* Are you trying to make money off this thing?

And I've been trying to train my friends to ask me things like...

* How would the world have to change for this thing to be a viable product?

* Do you really want to put in the effort for THAT when you could have the fun of building something totally new.

Which reminds me... if you're any way like me, part of your motivation is to just do something cool or new.

But... even smart, supportive friends aren't going to easily see your vision. (I tell every young engineer I meet "dude. ya gotta get good at explainin' stuff.") But hopefully they'll ask questions that make you think whether a particular idea is worth taking to the next step.

Oh hey... here's an idea... UnfinishedCon: the conference for half finished ideas and products. Where you come and pitch an idea you're working on and the culture is you can't say "oh. that's a dumb idea," and before you could say "I don't think I would buy that (or invest time researching it if it's an open source project)" you have to come up with a suggestion to make it more palatable to you.

I wound up being a jr. partner in one company that was sold. A principal in another company that was sold and I sold a QR thing to a marketing company (just sold the product and software, not the whole company.) I didn't get super-wealthy with any of these transactions, but got enough to go back to working on random weird things that seem to make me happy.

Don't build, talk to people. Find people who would actually pay money for your potential idea/product/tool. I remember a interview with Twillio's CEO where he described how he came up with the idea of an API for text messages, he was meeting with a developer over a different product, and pitched the idea, and at first that developer didn't seem interested, but 5 minutes later that developer started asking more questions where the idea of texting api was taking roots.

HN is definitely a good place to get brutal feedback on ideas.