<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">On 08/21/2014 12:34 PM, Dolph Mathews
      wrote:<br>
    </div>
    <blockquote
cite="mid:CAC=h7gUNPzeEpV3sLJZNVB89t+ptoaZ6YE55c6nmNNiR42W1LQ@mail.gmail.com"
      type="cite">
      <div dir="ltr">
        <div class="gmail_extra"><br>
          <div class="gmail_quote">On Thu, Aug 21, 2014 at 11:21 AM,
            Daniel P. Berrange <span dir="ltr"><<a
                moz-do-not-send="true" href="mailto:berrange@redhat.com"
                target="_blank">berrange@redhat.com</a>></span>
            wrote:<br>
            <blockquote class="gmail_quote" style="margin:0px 0px 0px
0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
              <div class="">On Thu, Aug 21, 2014 at 05:05:04PM +0100,
                Matthew Booth wrote:<br>
                > "I would prefer that you didn't merge this."<br>
                ><br>
                > i.e. The project is better off without it.<br>
                <br>
              </div>
              A bit off topic, but I've never liked this message that
              gets added<br>
              as it think it sounds overly negative. It would better
              written<br>
              as<br>
              <br>
                "This patch needs further work before it can be merged"<br>
            </blockquote>
            <div><br>
            </div>
            <div>++ "This patch needs further work before it can be
              merged, and as a reviewer, I am either too lazy or just
              unwilling to checkout your patch and fix those issues
              myself."</div>
          </div>
        </div>
      </div>
    </blockquote>
    <br>
    Heh...well, there are a couple other aspects:<br>
    <br>
    1.  I am unsure if my understanding is correct.  I'd like to have
    some validation, and, if I am wrong, I'll withdraw the objections.<br>
    <br>
    2.  If I make the change, I can no longer +2/+A the review.  If you
    make the change, I can approve it.<br>
    <br>
    <br>
    <blockquote
cite="mid:CAC=h7gUNPzeEpV3sLJZNVB89t+ptoaZ6YE55c6nmNNiR42W1LQ@mail.gmail.com"
      type="cite">
      <div dir="ltr">
        <div class="gmail_extra">
          <div class="gmail_quote">
            <div><br>
            </div>
            <div><a moz-do-not-send="true"
                href="http://dolphm.com/reviewing-code">http://dolphm.com/reviewing-code</a><br>
            </div>
            <div><br>
            </div>
            <blockquote class="gmail_quote" style="margin:0px 0px 0px
0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><br>
              as that gives a positive expectation that the work is
              still<br>
              wanted by the project in general<br>
              <div class=""><br>
                <br>
                > This seems to mean different things to different
                people. There's a list<br>
                > here which contains some criteria for new commits:<br>
                ><br>
                > <a moz-do-not-send="true"
                  href="https://wiki.openstack.org/wiki/ReviewChecklist"
                  target="_blank">https://wiki.openstack.org/wiki/ReviewChecklist</a>.<br>
                ><br>
                > There's also a treatise on git commit messages and
                the structure of a<br>
                > commit here:<br>
                ><br>
                > <a moz-do-not-send="true"
                  href="https://wiki.openstack.org/wiki/GitCommitMessages"
                  target="_blank">https://wiki.openstack.org/wiki/GitCommitMessages</a><br>
                ><br>
                > However, these don't really cover the general case
                of what a -1 means.<br>
                > Here's my brain dump:<br>
                ><br>
                > * It contains bugs<br>
                > * It is likely to confuse future
                developers/maintainers<br>
                > * It is likely to lead to bugs<br>
                > * It is inconsistent with other solutions to
                similar problems<br>
                > * It adds complexity which is not matched by its
                benefits<br>
                > * It isn't flexible enough for future work landing
                RSN<br>
                <br>
              </div>
              s/RSN//<br>
              <br>
              There are times where the design is not flexible enough
              and we<br>
              do not want to accept regardless of when future work might
              land.<br>
              This is specifically the case with things that are adding
              APIs<br>
              or impacting the RPC protocol. For example proposals for
              new<br>
              virt driver methods that can't possibly work with other
              virt<br>
              drivers in the future and would involve incompatible RPC
              changes<br>
              to fix it.<br>
              <div class=""><br>
                > * It combines multiple changes in a single commit<br>
                ><br>
                > Any more? I'd be happy to update the above wiki
                page with any consensus.<br>
                > It would be useful if any generally accepted
                criteria were readily<br>
                > referenceable.<br>
                <br>
              </div>
              It is always worth improving our docs to give more
              guidance like<br>
              ths.<br>
              <div class=""><br>
                > I also think it's worth explicitly documenting a
                few things we<br>
                > might/should mention in a review, but which aren't
                a reason that the<br>
                > project would be better off without it:<br>
                ><br>
                > * Stylistic issues which are not covered by HACKING<br>
                ><br>
                > By stylistic, I mean changes which have no
                functional impact on the code<br>
                > whatsoever. If a purely stylistic issue is
                sufficiently important to<br>
                > reject code which doesn't adhere to it, it is
                important enough to add to<br>
                > HACKING.<br>
                <br>
              </div>
              Broadly speaking I agree with the idea that style cleanups
              should<br>
              have corresponding hacking rules validated automatically.
              If some<br>
              one proposes a style cleanup which isn't validated I'll
              typically<br>
              request that they write a hacking check to do so. There
              might be<br>
              some cases where it isn't practical to validate the rule
              automatically<br>
              which are none the less worthwhile -1'ing  for - these
              should be the<br>
              exception though. In general we shouldn't do style
              cleanups that we<br>
              can not automatically validate in some way.<br>
              <div class=""><br>
                > * I can think of a better way of doing this<br>
                ><br>
                > There may be a better solution, but there is
                already an existing<br>
                > solution. We should only be rejecting work that has
                already been done if<br>
                > it would detract from the project for one of the
                reasons above. We can<br>
                > always improve it further later if we find the
                developer time.<br>
                ><br>
                > * It isn't flexible enough for any conceivable
                future feature<br>
                ><br>
                > Lets avoid premature generalisation. We can always
                generalise as part of<br>
                > landing the future feature.<br>
                <br>
              </div>
              See my note about - it isn't always just about premature
              generalization<br>
              per se, but rather seeing things we are just clearly
              written from too<br>
              narrow a POV and will cause us pain down the line which
              could be easily<br>
              mitigated right away.<br>
              <br>
              Regards,<br>
              Daniel<br>
              <span class=""><font color="#888888">--<br>
                  |: <a moz-do-not-send="true"
                    href="http://berrange.com" target="_blank">http://berrange.com</a> 
                      -o-    <a moz-do-not-send="true"
                    href="http://www.flickr.com/photos/dberrange/"
                    target="_blank">http://www.flickr.com/photos/dberrange/</a>
                  :|<br>
                  |: <a moz-do-not-send="true"
                    href="http://libvirt.org" target="_blank">http://libvirt.org</a> 
                              -o-             <a moz-do-not-send="true"
                    href="http://virt-manager.org" target="_blank">http://virt-manager.org</a>
                  :|<br>
                  |: <a moz-do-not-send="true"
                    href="http://autobuild.org" target="_blank">http://autobuild.org</a> 
                       -o-         <a moz-do-not-send="true"
                    href="http://search.cpan.org/%7Edanberr/"
                    target="_blank">http://search.cpan.org/~danberr/</a>
                  :|<br>
                  |: <a moz-do-not-send="true"
                    href="http://entangle-photo.org" target="_blank">http://entangle-photo.org</a> 
                       -o-       <a moz-do-not-send="true"
                    href="http://live.gnome.org/gtk-vnc" target="_blank">http://live.gnome.org/gtk-vnc</a>
                  :|<br>
                </font></span>
              <div class="">
                <div class="h5"><br>
                  _______________________________________________<br>
                  OpenStack-dev mailing list<br>
                  <a moz-do-not-send="true"
                    href="mailto:OpenStack-dev@lists.openstack.org">OpenStack-dev@lists.openstack.org</a><br>
                  <a moz-do-not-send="true"
                    href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev"
                    target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
                </div>
              </div>
            </blockquote>
          </div>
          <br>
        </div>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
OpenStack-dev mailing list
<a class="moz-txt-link-abbreviated" href="mailto:OpenStack-dev@lists.openstack.org">OpenStack-dev@lists.openstack.org</a>
<a class="moz-txt-link-freetext" href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a>
</pre>
    </blockquote>
    <br>
  </body>
</html>