* Alon Bar-Lev <alo...@redhat.com> [2012-10-02 10:26]:
> 
> 
> ----- Original Message -----
> > From: "Ryan Harper" <ry...@us.ibm.com>
> > To: "Alon Bar-Lev" <alo...@redhat.com>
> > Cc: "Ryan Harper" <ry...@us.ibm.com>, "Doron Fediuck" 
> > <dfedi...@redhat.com>, "Mark Wu" <wu...@linux.vnet.ibm.com>,
> > "Greg Padgett" <gpadg...@redhat.com>, vdsm-devel@lists.fedorahosted.org, 
> > "Ayal Baron" <aba...@redhat.com>, "Dan
> > Kenigsberg" <dan...@redhat.com>
> > Sent: Tuesday, October 2, 2012 5:06:28 PM
> > Subject: Re: Change in vdsm[master]: Use 'yum clean expire-cache' instead 
> > of 'yum clean all'
> > 
> > * Alon Bar-Lev <alo...@redhat.com> [2012-10-02 09:43]:
> > > > From: "Ryan Harper" <ry...@us.ibm.com>
> > > > To: "Dan Kenigsberg" <dan...@redhat.com>
> > > > Cc: "Alon Bar-Lev" <alo...@redhat.com>, "Doron Fediuck"
> > > > <dfedi...@redhat.com>, "Mark Wu" <wu...@linux.vnet.ibm.com>,
> > > > "Greg Padgett" <gpadg...@redhat.com>,
> > > > vdsm-devel@lists.fedorahosted.org, "Ryan Harper"
> > > > <ry...@us.ibm.com>, "Ayal
> > > > Baron" <aba...@redhat.com>
> > > > Sent: Tuesday, October 2, 2012 4:32:30 PM
> > > > Subject: Re: Change in vdsm[master]: Use 'yum clean expire-cache'
> > > > instead of 'yum clean all'
> > > > 
> > > > > > 
> > > > > > Hi Dan,
> > > > > > 
> > > > > > I don't understand why you don't treat "0" at the above...
> > > > > > 
> > > > > > If there were no worries, +1 had been provided...
> > > > > 
> > > > > To me, "0" means "I do not have a strong opinion, I trust other
> > > > > people
> > > > > to make the right decision, given the facts and worries that I
> > > > > have
> > > > > raised".
> > > > > 
> > > > > Sometimes I do not give a +1 simply because I did not have time
> > > > > to
> > > > > review
> > > > > the whole code, not because I have a strong worry. "0" means
> > > > > "not
> > > > > reviewed yet" or "not endorsed yet by me but not rejected by
> > > > > me".
> > > > > 
> > > > > A polite and well-detailed -1 should be used daily and not
> > > > > considered
> > > > > "rude".
> > > > 
> > > > Agreed.  It took a little getting used to for myself (mostly
> > > > because
> > > > of
> > > > the 'I would prefer you didn't submit this'), but I think a -1
> > > > with comments is much more valuable than a 0 with comments. If I
> > > > leave a
> > > > 0 with comments, I feel that I'm telling the maintainer that I
> > > > don't
> > > > feel strongly enough to force the submitter to change the code.
> > > > 
> > > > Using -1 or +1 is similar to Ack and Nack, which give an
> > > > affirmative
> > > > yes or
> > > > no.  Should this patch be merged in its current form?  That's a
> > > > binary
> > > > question, hence the need for a +1 or -1.
> > > 
> > > I disagree in that regard.
> > > 
> > > There is a huge difference between discussion and rejection.
> > > Having the scale of -1, 0, +1 and not using 0 for discussion
> > > actually narrow the scale by 30% to -1, +1.
> > > 
> > > ACK - is a complete ACK to be merge "If I was the maintainer I
> > > would have submitted".
> > > NAK - is a complete NACK "If I was the maintainer I would have
> > > rejected".
> > > 
> > > ACK/NAK are methods of decisions for the entire patch, not for
> > > iterative discussion, nor for bookmark what I've reviewed.
> > > 
> > > Maybe in future the scale will be larger... -3...+3, so
> > > unprivileged user get -2..+2, then minor comments can be at -1..+1
> > > range, but we are not there.
> > > 
> > > I don't like we narrow the scale in vdsm in 30%...
> > 
> > I don't care that much for a scale; the numbers are arbitrary IMO.
> > 
> > I just need a way to ensure that the maintainer (and the submitter)
> > understand that that IMO, the current patch isn't ready for
> > acceptance.
> > And that the objection to submission in the current state isn't lost.
> > If I provide review comments, but don't include a -1, it wasn't clear
> > to
> > me that the maintainer would view this comment as a real concern.  A
> > -1
> > plus constructive review comment IMO makes this point clearly to the
> > maintainer.
> > 
> > Dan has clarified his view on the scale; and as long as I understand
> > how
> > it's being used; that's good enough for me.
> 
> It is just harder to work this way, this applies Dan as well who
> abuses the -1, and applies only to vdsm among all the ovirt
> repositories.
> 
> I will repeat last time as I still unsure that the argument was clear
> enough, no need to discuss this any farther.

You were perfectly clear; I just disagree.

> 
> You, as contributer that do not have -2, +2, have no mean to
> differentiate your view between "I don't like this patch to be
> included, no matter what minor issues are resolved", and "I find this
> patch to be useful and I work toward making it perfect", when it is
> perfect I clearly state so using my +1 vote.

I have no issue marking +1 when I'm happy with the patch.  What I want
is way to raise a flag to the maintainer that I have a concern I would
like addressed before the patch is merged.  If the maintainer chooses to
merge a patch even though a reviewer has marked -1, that's fine.  But I
don't think just a 0 + comments is a strong enough position.  Or at
least, I wanted clarification from the maintainers for their
preference.

If we didn't collapse a discussion into singular numbers, we wouldn't
wouldn't be discussing what a -1, 0, or +1 means.  The words we've
written would matter more.  



-- 
Ryan Harper
Software Engineer; Linux Technology Center
IBM Corp., Austin, Tx
ry...@us.ibm.com

_______________________________________________
vdsm-devel mailing list
vdsm-devel@lists.fedorahosted.org
https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel

Reply via email to