Hi Gary,
On 03.01.2013 00:50, Gary Mills wrote:
On Wed, Jan 02, 2013 at 04:38:44PM -0500, Gordon Ross wrote:
On Wed, Jan 2, 2013 at 3:16 PM, Milan Jurik <milan.ju...@xylab.cz>
wrote:
>
> It would require to have somebody pushing it to Illumos, with at
least 2
> reviewers. My current experience is not very good in pushing
larger
> things there, ext2fs did not find any reviewer for a year, WPA
> enterprise implementation with WiFi improvements and clean ups are
also
> waiting.
>
> In case of FUSE I am not even trying because at least one highly
> respected RTI advocate was strong opponent of this FUSE
implementation
> even during his SUN carreer.
I hope you're not referring to me above.
I'd like to see these move ahead,
but have not had time to review...
The problem seems to be, and I'm guilty of this myself, is that too
many people are working on development and too few people are doing
reviews. In many cases, I don't feel competent to review a change
because I'm not familiar with the area in question.
Is the code review supposed to examine only the quality of the code,
or is it also supposed to determine if the code will actually
function
correctly? I can generally do the former, but often not the latter.
If the submitter includes information on the test framework they
used,
is that all that's needed for the function of the code?
The problem with missing reviewers existed for all my SUN life and
probably even before that. Nobody wants to "waste" his time on something
he has no own interest (and sometimes even if he has). If we will wait
for 2 experts in area then we will never touch many places in code. Even
in SUN it was hard to find so many people with expertise of some parts
of system. But I cannot criticize people, I do not spend enough time on
reviews these days. TODO list is too long and my time is too short for
Illumos/OI. Sorry.
Reviewer doing evaluation of the quality of the code is frequently much
better than expert reviewer with no time. I am not RTI advocate so I
cannot decide how good must reviewer be, of course.
Best regards,
Milan
_______________________________________________
OpenIndiana-discuss mailing list
OpenIndiana-discuss@openindiana.org
http://openindiana.org/mailman/listinfo/openindiana-discuss
