Lorin Hochstein 
What's the next hype cycle after AI?
Jenniferplusplus@norootcause well, we had cloud, then kubernetes, then serverless, then magic fake digital money, then magic computers that can do anything. I don't know how to escalate from there. Maybe single purpose on-prem appliances? Just collect $200 and start over at go?
ghaff@jenniferplusplus @norootcause I wonder if there's some post-Kubernetes thing but I wonder if that's like post-Linux at this point. For the foreseeable future, largely incremental rather than the wholesale shift (notwithstanding antecedents( that played into containers, Kubernetes, and cloud-native generally).
@jenniferplusplus Linux-based unikernels could be part of that but I remember the last flameout of server appliances. Unikernels, if they catch on, would mostly be an evolution of the Linux process model.
@ghaff @norootcause Maybe double down on distributed/edge computing, and try to push virtually all of the compute out onto user devices?
Or not, since all the hype cycles seem to be things that are good for MS, amazon, and/or google.
@jenniferplusplus Edge is certainly big already although maybe not at the top of the hype cycle. I do think figuring out where computing (and storage) belong for different use cases is an important debate for some number of upcoming years.
@jenniferplusplus It's *somewhat* evolutionary but some combination of the evolution of distributed computing, processors/computing architectures, what runs on those processors/computing architectures, etc. are all going to be important even if fragments of what will happen are almost certainly visible today.