Currently in planning phase and searching for volunteers, some of the features we hope to achieve are:

Focus on open standards.

Ability to import and export from most, if not all, of the widely accepted genealogical data exchange formats (GEDCOM, GENTECH and the various XML flavors, Family Tree Maker and other native program formats).

Able to manage large amounts of data quickly and efficiently.

Easy to use, intuitive interface that facilitates a quick understanding and shallow learning curve for professional and amateur genealogists alike; a focus on getting the job done instead of requiring vast amounts of computer knowledge to configure and run.

Source indexing and application system that allows users to quickly create a source based on predefined templates (i.e., newspaper article), upload a graphic image of the source, and drag and drop that source to a fact that might exist.

Similar indexing and application system for media. Quick addition and indexing of family photos and the ability to drag and drop a photo to multiple individuals. If I had to give an example of this functionality, it would be Adobe's Photoshop Album program.

Eventually, a lofty goal would be porting the application to a "lite" version and offering it for use on handhelds such as Palm, Windows CE, or iPhone for researchers on the go.

Online publishing and community development, something like PHPGedView?

A focus not on creating a better product than other solutions, but rather developing a tool that works in turn with other applications based on open standards. It's quite possible that the vision for Treefrog will merge with or disappear in favor of another open source project out there; this product is not important, but the end solution of a product that accomplishes these goals is essential for genealogists.