Source code management history browser

Often, how a piece of code arrived at it's current state is more important than what the code does. In my experience, the history of how a piece of code evolved is stored organizationally in the minds of those who have been working with a code for quite some time. This system is inefficient, and leads to unnecessary mistakes due to loss of information.

On the other hand, we have rich source code managment systems where every change to the source code is commented, sometimes with rich descriptions. However, also in my experience, this rich history is never used, or only looked at when something goes wrong.

I would like to develop a system that unlocks this history for programmers. I envision an eclipse plugin where you can highlight a block of code and view all the changesets relevant to how the code got that way, along with viewing changes to related files. I think that such a program would greatly enrich maintenence programmer's understanding of the code.

An alternative but similar project would be to collect usage statistics for many closed and open sourced centralized repositories, and look at how the repositories are used. How often to programmers look at something done 2 weeks ago, for example? How about one week?

search terms: configuration management, software evolution

Darcs: Distributed version management in haskell

Visual support for version management

SnapChain: a shared snapshot method for data version management

A case-study of configuration management with ClearCase? in an industrial environment

What software evolution and biological evolution don't have in common

Challenges in software evolution

Evolution storyboards: Visualization of software structure dynamics

Characterizing programmers' information-seeking during software evolution

Software bugs and evolution: a visual approach to uncover their relationship

Studying the evolution of software systems using change clusters

Evolution and growth in large libre software projects

as far as I can tell, no research into how repositories are actually used.