I've been re-evaluating lately how many cloud services I'm using. I've noticed that without realising it pretty much everything I have relies on a handful of companies' servers.
This didn't always used to be the case either. Back in 2011/2012 I was happily using a home server to run an instance of iTunes and Plex Media Server for sharing music, files, tv shows, etc. Nowadays though I use Netflix and Spotify. Before GitHub came along my home server would be my Git server, now all my code is stored on someone else's machine.
As cloud services become more ubiquitous, data ownership is becoming a strange and foreign concept. However, with that comes the risk of losing access to something you love from a change in law, licensing agreements, or simply the bankruptcy of a company. I've experienced this myself - losing access to music because Spotify didn't strike a new agreement with an artist.
I've decided to privatise and regain control over the most important things. My personal projects and my music collection are the main focus, and my blog is now quite 'old-world' in that it's RSS-only, nothing gets put on social networks.
I feel that data ownership is quite an important thing, and it's something we've almost completely lost in the advent of cloud-everything. Whilst cloud services offer a lot of convenience, you're ultimately at the mercy of a company that cares very little about your interests, and you stand to lose everything if they decide to cease operating.
In terms of my music collection, I'll probably go back to using Airsonic, and for my personal projects I've moved everything to Phabricator, which is like JIRA and GitHub merged together. Both run on the home server in my cupboard.
It feels great to have that control over my data again.