I'm still in the startup phase, and the time saving of Heroku has been fantastic. I'm ever mindful of these types of stories and the scaling cost though. I've

I'm still in the startup phase, and the time saving of Heroku has been fantastic. I'm ever mindful of these types of stories and the scaling cost though. I've not seen it mentioned so far: RedHat has OpenShift (https://www.openshift.com/ https://www.openshift.com/) out now, v3 will see Docker support too. With only a quick play, it seems to be a mix of Heroku and dedicated host. On sysadmins: talking to a guy who managed AWS nodes for a popular iOS game you've probably played, he made the comment that it was great to have a new team member onboard now, dedicated to looking after all the VMs. I figure he's rediscovered sysadmins, just using different kit. The risk I see with managing your own Dokku, etc is handling issues/maintaining uptime if it's not your only job/skill. When another HeartBleed comes along, can you react quickly and competently enough?