Martin Schaffstall wrote:
It is interesting to observe that each time the project reaches
another milestone a Sun employee steps in and demands a full change of
the project.
Interesting...
As an observer who is keenly interested in ksh93 matters, but has not
spent time working on the details of *this* project, I interpreted this
discussion as:
The project team *assumed* a certain set of constraints (cost of
flag days, library names...)
and designed their project around them.
When the project got to a significant milestone (Yea!!!) and was
ready to go on
to the next one, they found out (#include DP's email here...) that
some of their
assumptions were not as set in stone as they thought and others lead
to undesirable
outcomes.
This is both good and bad.
While not many people like to see their assumptions invalidated, it is
good that this happened _now_ rather than after the project thought it
was completely finished; it is bad that it did not come up even
_earlier_, before these assumptions found their way into the design and
implementation.
Yes this is a setback, but think it through: If these issues had not
been discovered now, and the project had continued as is, the cumulative
negative impact on future work (integration, maintenance,
portability...) almost certainly would have been worse.
I'd hate to misread your remarks as saying that we should ignore
incomplete architecture or design simply because we didn't discover gaps
until after the implementation was complete, or that it is anything but
good software engineering practice to take time to understand and work
through the architectural implications of problems as they arise.
-John (who has a bias towards proactive architectural efforts) Plocher
_______________________________________________
opensolaris-discuss mailing list
[email protected]
