I recently ran across gource, which shows the development of a project over time by looking at the commits in a version control system. I pointed it at our CVS server, and this is what it spit out: http://www.youtube.com/watch?v=ct1tIoLJPjs
I thought it was interesting that the programming activity hit the high point around week 6, and stayed strong all the way to the championship.
Very cool. Are the dots meant to represent files? What causes some of them to branch off?
Each dot is a file, and each branch is a sub-directory.
Seeing that you use LabVIEW, how do you handle merging VI’s?
We typically tried to avoid merging. We tried to work on different parts of the code, so that people wouldn’t have to merge. There were times where someone delayed implementing a change until someone else had finished with that VI. When we failed to communicate and ended up needing to merge, we used the LabVIEW merge tool, which most of the time worked fine.
One thing we noticed is there are times when there were unexplained changes (someone didn’t explicitly change a VI, but it changed anyway). We were able to minimize these by turning on the separate compiled code from VIs. We also made each global variable its own VI, because we found that changing anything in the global variable VI made LabVIEW recompile every VI that uses that global variable VI).
I think the more teams that see this the more teams will make better use of version control :D.
In all seriousness, wow, I’m jealous I’m going to use this for the team and personal projects :).
I ran the gource visualization again for our code this past year. http://www.youtube.com/watch?v=OBuoa5eqmbM
If that was our screen, it would have been completely black until the day of our first regional. Our code just didn’t work (for reasons unknown) when we switched cRIOs (from practice to comp bot) plus we (stupidly) forgot joysticks and had to buy new ones (different). Our lead programmer was up all night Thursday and we played the 1st match Fri without a hitch
This is very cool! I think it gives a very good representation of what actually takes place over a season.
I wonder if teams could use this as a visualization on mapping out what has been done in code and what needs to be done next.
I really like this.
Thanks!
I just compiled and ran this on our repos for the last two years, that was absolutely amazing!
Thank you for sharing this great find!