[ansonparker]

This approach is very common. Any start-up dismissed as "a feature, not a product" has pretty much started this way: 1. take a product you like, but that is missing a feature you really want; 2. clone with added feature; 3. launch.

More often than not this approach fails either because of the high cost of migration to potential users or more likely the pet feature doesn't actually appeal to any more than a small niche.

Don't think this necessarily applies to Google in this case. But I think Google can pull together a user base of 10/20 million pretty easily. The success of Google+ will be determined by whether they can get into the 100's of millions.