On Sun, Feb 14, 2010 at 10:03 AM, Sean Owen <sro...@gmail.com> wrote:
> It would be far too much overhead at this stage to manage every change with
> a patch. If in any reasonable doubt about a patch or whether someone would
> object, yes open JIRA.
>

A couple of points. I am +1111111111111 to the appropriate use of lazy
consensus. I think that I was overly worried about  accidentally
applying my CXF habits at Mahout. So if some of you have been
wondering why I have seemed to be a constant advertisement for my
patches, now you know why.

I am personally somewhat squeamish about public final methods. To me,
there's a sort of Murphy's Law that applies. As soon as you release
with a public final method, someone will come up with a legitimate
reason to override it. I've been very frustrated in the past by
over-enthusiastic finalizers.

I have no special opinions about final fields. My comments earlier
were to the effect that I don't find them (outside of statics) wildly
useful, but I also don't mind them and won't object if Mahout
checkstyle/pmd decided to force me to use more of them.

Reply via email to