App sandboxing is… the way it will go, for insurance reasons.
App sandboxing is… the way it will go, for insurance reasons.
At least one big company already maintains internal clones of repos with registration of every version of every component used (I know this through personal experience) and I have no doubts the others are doing the same, especially if they are into cloud or government contracts.
Do these companies have people on-call who can security review urgent security fixes in the components? Or how does it work
(say, Log4j and the fixes need to get reviewed urgently, and let's pretend the language wasn't Java but a language that fewer people could understand?)
You're saying that like it is a bad thing?
A developer probably shouldn't have access to all that live data anyways.
Unfortunately unless you get ridiculously creative and put in a bunch of extra time instead of cloud service spend, SOC2 compliance is gonna be expensive
For this kind of issue one could use VM's where you encrypt the VM with the live customer data. Another option would be to use cloud.
But yes.. there's also nothing wrong with having more than one machine to physically separate the systems and thus data. Bonus on that one is that you can shut it down if not needed and as such it won't be accessible over the network unless switched on.
Personally -as a small developer myself- I use all of the above.