<div dir="ltr"><div class="gmail_extra"><br><div class="gmail_quote">On Mon, Oct 7, 2013 at 4:27 PM, Jim Grosbach <span dir="ltr"><<a href="mailto:grosbach@apple.com" target="_blank" class="cremed">grosbach@apple.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Good thoughts, Chandler. Do I understand you correctly that you agree with Evan’s statement with the modification of s/committed/enabled by default/?</blockquote>
</div><br>Essentially, although I tried to be a bit more clear about which buckets different optimizations fall into. For example, in this case, I don't think this change would would have fallen into the bucket where you really need to give some significant heads up prior to enabling by default.</div>
<div class="gmail_extra"><br></div><div class="gmail_extra">I think the problem is this case was very different -- the people making the change *knew* there were a small number of important benchmarks that regressed, and failing to address that prior to enabling or committing something is quite different.</div>
</div>