<div dir="ltr"><div class="gmail_extra">On Fri, Nov 15, 2013 at 11:45 AM, Brian Wood <span dir="ltr"><<a href="mailto:woodbrian77@gmail.com" target="_blank">woodbrian77@gmail.com</a>></span> wrote:<br><div class="gmail_quote">
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><div>Not all<br></div>clouds are helpful though. They've spent over 600 million<br>
on <a href="http://healthcare.gov" target="_blank">healthcare.gov</a> and they're spending tons more now to<br></div><div>"fix" it.</div></blockquote></div><div class="gmail_extra"><br></div>That has absolutely *nothing* to do with the cloud, but rather lack of diligence in development, test, and QA.</div>
<div class="gmail_extra"><br></div><div class="gmail_extra">I just wrapped up attending the AWS re:Invent conference, where ~9k people paid big money to come and learn better development, architecture, and deployment techniques. If the cloud were magic, that wouldn't happen. Though cloud architecture strategies look quite different than "legacy" systems, a large portion of the skills carry over.</div>
<div class="gmail_extra"><br></div><div class="gmail_extra">In fact, IMHO the patterns used to build highly-available, high-performing applications in the cloud (whether a local cloud, Rackspace, AWS, whatever) actually would be beneficial to employ when building bare-iron systems as well, due to the extreme focus in eliminating SPOFs.</div>
<div class="gmail_extra"><br></div><div class="gmail_extra">-Erik<br><br></div></div>