CS 481 - Team 49
Meeting Minutes

Weekly Team Meeting
Friday, October 12, 2007
10:30 - 11:30 AM
JEB 324
Recorder: Brandon Riggers
--------------------------------

Meeting Summary
   This meeting involved discussion with Bruce and Bill 
about our goals for release #2, and how our tool will 
interact with Cruise Control.  Bruce also expressed a 
slight "anxiousness" with the amount of work and discovery  
we still have to do, as well as our team's shared 
understanding.  Toward the end, Bruce also talked about 
how someone at HP asked him earlier that morning about 
our tool, and that our tool will indeed be used.
   After the teleconference portion with Bruce was over, 
Bill expressed the need to add (and carry out) design-
oriented tasks on our task list, and reiterated the idea 
of communicating more with Bruce.  Travis was present 
toward the end of the meeting, and got a summary of 
what we discussed.


Attendees
   Brandon Riggers, Peter Brown-Hayes, Bill Junk, 
Bruce Mayes (over the phone), Travis Thoroman (for last
portion)


Topics Discussed
   The meeting began with Bill briefly asking if we had 
run any accuracy or performance tests.  Peter then went 
over the contents of release #2, which will be the entire 
run_turmoil package.  This will include interfaces with 
the command line and Subversion, but will not include the 
turmoilBetweenRevisions portion of the tool.
   The next topic involved discussion with Bruce about 
what we can expect from Cruise Control.  Specifically, 
Peter asked what our tool will get from Cruise Control 
when a commit is done.  Bruce said he expects a turmoil 
summation to be at the end of the automated email, and 
that Cruise Control will invoke our tool and give it the 
revision numbers.  A single file in a commit being at a 
higher revision number than other files should not effect 
our tool.  Bruce stated that we should just get everything 
that has changed, and Bill added that on some level, we 
could be ignoring the individual file names (or 
revisions). Bruce also talked briefly about how you can 
commit on a branch in Subversion, but Cruise Control will 
not know it (or, react).  He suggested that in this case, 
the user may call our tool manually.
   Bruce then expressed his feeling of "anxiousness."  He 
mentioned that our first release was good, but noted that 
a "fair amount" of work and discovery remain.  He is a 
little uneasy about the one file that failed in our tests. 
Bruce also talked about our shared understanding of the 
tool, and that our project is a little different than 
those in past CS481 classes.  Bill suggested that we at 
least begin givng Bruce a mid-week email to communicate 
our status - simply to keep up communication and let 
Bruce know how things are going.
   Bruce lastly told of how someone at HP asked him that 
morning about this project.  Bruce said there is a 
definite interest on that end in integrating the tool, 
and the goal is to use it in their environment.  Our 
tool will be used, and an audience exists for it.
   After the teleconference, Bill mentioned how we needed 
to add "designing" tasks to our task list, and how we 
should be designing the release, and then implementing 
it.  Bill also re-stated the need to communicate with 
Bruce, and even task Bruce with certain things or ask 
him for help (like with Subversion or Cruise Control). 
Bill lastly noted that we can sense that our tool is 
practical.
   Travis joined us toward the very end of the meeting, 
when he was filled in with a summary of the discussion.