On Wed, 11 Jul 2001, KUMAR,KRISHNA (HP-Cupertino,ex3) wrote:
> If a new baseline is created on top of the old baseline, still you can
> move the labels in the old baseline and everything looks fine, even if
> the new baseline is used by other views.
> Though I would love to see, rational come out with a feature, in which
> I can make a new baseline from any baseline I wanted rather than from
> the latest baseline. This way a bug can be fixed and create a new
> baseline, without affecting the recent baseline stuff. However
> Rational Tech support says it is highly unlikely to going to happen.
This is an example of the false expectations generated by the baseline
concept.
In practice, the only meaningful baseline _is_ the latest baseline. The
idea that one should be able to pick and choose baselines and somehow the
right build will magically appear is _wrong_. There is no magic.
Yes, you can solve your immediate hurt by twiddling with the labels, but
if someone makes a bad delivery, the only meaningful way to fix it is by
doing _another_ delivery that either fixes or undoes the change. Moving a
label backwards will only postpone the problem and create confusion, since
now the association between the baseline and the change sets is no longer
valid. In other words, people will run into the "hidden" change at their
next delivery and wonder why there is a conflict, or simply the next new
baseline will again uncover the "hidden" change and create a "regression".
See also http://www.miaow.com/clearcase/faq/ucm.html
-- cg- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
This archive was generated by hypermail 2b29 : Tue Jul 31 2001 - 22:03:53 EDT