Administrators of organization computers are often very careful about what they allow to be installed on computers. They often disable automatic updates, and instead vet the updates manually and push them to computers if they are sure it didn't break any user workflows.
But then, comes in Chrome, which has tendency to update itself automatically and frequently. I don't know if Chrome provides a "gated" update, that allows organization administrators to test the updates before letting it inside the organization. But I can imagine it does.
My question is this. How are applications, which update themselves automatically, frequently and often without user even knowing, seen by administrators of organization computers? If I were to create such an application, would those administrators require ability to turn off the auto updates? Or would they be fine with being able to check the updates before hand?
I realize this question might be too broad and depend heavily on specific organization. I'm more interested in "best practices" and "general rules" answer.
The reason why I'm asking is that idea of Continuous Deployment has been growing strong in software development. This is much easier with server components, when you control the server the software is deployed on. But it becomes difficult with software installed directly on client stations. I believe Chrome is doing Continuous Deployment, which is why it is relying heavily automated updates of new releases. Those can, in extreme cases, be daily incremental builds. What I want to know what kind of friction I should expect from administrators if I were to sell application, that tries to update itself weekly or even daily? And possibly what should this kind of automated update system provide to administrators to make it easier for them. As I wouldn't want admins go against the application if it breaks their update workflows.