Hoping that someday soon^tm someone will come up with something that makes all the current mess go away is a dangerous path.
In germany they call it 'Technologieoffen' - and even some political parties have written it on their flags.
'No need to implement changes now - someday God will decent and absolve us from all of our sins - until then party hard! #yolo'
I'm all for accelerating research/tech. But it can't be an excuse to just let sh*t happen.
Yet most people will do this - it's expected human behavior.
Mine still runs happily on a PI Model B Plus with no downtime except for maintenance.
Some DNS/Netwrk fallbacks on the DSL Router.
Imho the risk/consequences of a downtime don't justify having some hardware draining power.
If all goes down ansible will have provisioned a new pi in no time.
Nevertheless i applaud the effort and am happy to have learned some things from the article
Spinning up a server and installing a repo on it is easy. Depends on your use case and on what you know/have.
I prefer ansible or jenkins+scp-build-to-server+run-deploy.script
I added it to my tools list in case i need sth quick'n working for a small team/to recommend when there's no ansible/sysadmin knowledge available.
(I haven't looked into piku but i guess you'll hit its limitation once you have more complex deployment schemes, privilege/access management, ...)