On Tue, 20 Aug 2013 10:38:02 -0400, Konstantin Berlin wrote:
We already discussed that some time ago.
Let's not mix different issues. Removing the handling of weights can
dealt
with later.
I disagree, please address my points in regards to this, rather than
saying a generalizing statement.
Oh, you sure can disagree.
Having weights be included makes it harder to implement a unifying
API. In addition there is performance cost because the user supplied
Jacobian is adjusted.
Code welcome, with usual requirements: unit tests, and examples of how
to "externally" handle weights.
If it can be handled within the current design, this could be
incorporated independently of the broader issue which Evan proposed
to tackle.
But it will require the "go ahead" of people who refused to remove
the weights from the current API when the issue was last time raised.
So , I suggets that you ask the question explicitly, on this list,
and if the answer is positive, that you record in a new issue on
the bug-tracking system (with a link to the thread stating the
eventual agreement).
Regards,
Gilles
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org
