On 02/20/2012 10:34 PM, Samuel Greenfeld wrote:
In general I don't want to have test cases scattered all over the
place, as Sugar information already is found in at least three Wikis,
with developers only tending to update one or two.
(An interesting take on this can be seen in the opening speech of last
year's Google Test Automation Conference, titled 'Test Is Dead' -
http://www.youtube.com/watch?v=X1jWe5rOu3g . My personal view is that
while the extremes are interesting, I've read books talking about
both, and the truth is somewhere in-between.)
That video brings nothing new to the table from my perspective.
In the 21 century you just throw your idea/project/product over the wall
and in the hands of the users/consumer and they themselves take care of
testing and take direction of the project/product that is if they deemed
it not failed to begin with.
Today it's more about are you fast enough to respond to them (
users/consumer ) but mostly are you fast enough to respond faster then
your competitor.
( by responding I mean fix detected bug and implement what ever the
change the user needs/wants )
I dont know if the above makes sense to you.
In a project like Fedora which consist of multiple projects which in
essence are all thrown over the wall the above does not apply since the
rules are quite different.
We have to make sure all the projects play nicely to each other ( well
nicely enough to be shipped ) so yes we are somewhere between the "old"
model and the "new".
JBG
--
test mailing list
test@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe:
https://admin.fedoraproject.org/mailman/listinfo/test
