I never view closing down a project as a failure, if I make sure and record information about the project throughout its life. Then I have access to what was learned over its life. Good or bad.
I have many crazy ideas, executed in a variety of ways. All of them I learn something, otherwise I wouldn't do them.
I was just reading a post from
Google about the decommissioning of Google Health and Google PowerMeter. Even though I may not always agree, I am impressed by Google's ability to roll out new projects, then close them down when they haven't met their goals.
A great example of this was
Google Wave. Google Wave was innovative, but as a consumer product, it just didn't fly.
Apparently neither did
Google Health and
Google PowerMeter. These two areas represent such massive technology problems, I don't even know how you'd measure success. All you can do in these areas is try, try, and try again, and make sure you record data along the way, and make accessible when done.
Google has done a good job of making data from Google Health and Google PowerMeter available to individual users. It would also be helpful if Google open-sourced any technology, process and data from these projects as well.
Just like with Google Wave,
individual parts and pieces of a dead project could find a new life in other shapes and forms. Sometimes success for a project does not mean individual project success, it might be to establish a precedent, build knowledge or technology that can be used in other projects.