The folks at cloud9 IDE or Codecademy are in a good position to collect and expose patterns in programming efficiency.
On one's own, simply using a VCS with a high enough commit frequency would let you infer something about lines added per hour. Perhaps an editor's undo stack could also be committed to store finer-grained metrics.
With detailed metrics and impressive data analysis there's the opportunity to uncover useful insights into how we debug and learn. That's probably more than a weekend project, but worth it in potential future payout IMO.
Turns out I'm already collecting this data… Vim has recently added a persistent undo feature (enable by setting 'undodir' and 'undofile'; 'undolevel' can also be raised from the default 1000); there is a Gundo plugin that displays the undo tree like a DVCS commit graph.
The probability of such a program being developed is diminished by how depressing its results would likely be to its own author. It would be interesting, though, wouldn't it... now you've got me thinking.
zed shaw's peepcode one on one goes into something similar. By hooking into the make command, he builds up a nice stats database of similar information
Comments
Are there IDE's that track lines of code edited/added per hour, etc?
The folks at cloud9 IDE or Codecademy are in a good position to collect and expose patterns in programming efficiency.
On one's own, simply using a VCS with a high enough commit frequency would let you infer something about lines added per hour. Perhaps an editor's undo stack could also be committed to store finer-grained metrics.
With detailed metrics and impressive data analysis there's the opportunity to uncover useful insights into how we debug and learn. That's probably more than a weekend project, but worth it in potential future payout IMO.
Turns out I'm already collecting this data… Vim has recently added a persistent undo feature (enable by setting 'undodir' and 'undofile'; 'undolevel' can also be raised from the default 1000); there is a Gundo plugin that displays the undo tree like a DVCS commit graph.
The probability of such a program being developed is diminished by how depressing its results would likely be to its own author. It would be interesting, though, wouldn't it... now you've got me thinking.
I remember someone doing an experiment saving git branches every time he saved in his editor, which would give you some estimate into this:
http://www.jukie.net/bart/blog/save-everything-with-git-wip
But it doesn't look like he describes the result of this experiment.
zed shaw's peepcode one on one goes into something similar. By hooking into the make command, he builds up a nice stats database of similar information