Background & Motivation

My history with Linux and with Fedora (and CentOS) is pretty dense and somewhat intense going back to the first announcement of “Fedora Core” by Red Hat. And it extends to a short period of my professional career when I worked for Red Hat (and a much longer part of my career where I had a pivotal role as an individual contributor in IBM’s Linux Technology Center).

But my history with Apple goes back to about 1984, when I had saved up as a child and purchased an Apple IIe as my first computer.

I mention all of this because it’s hard to really contextualize my motivations without at least mentioning these factors, and underscoring why even considering this move is kind of a big deal for me.

Fundamentally, the catalyst to me exploring this is the really disappointing phenomenon where Silicon Valley’s billionaire class are tripping over themselves to suck up to convicted felon, Donald Trump. Not only are they making it clear the aren’t to be trusted with important roles in our digital lifestyles, but ethically I think it’s important to start to starve them of our discretionary spending. I’m calling this my deFAANG effort (where “FAANG” is a common acronym in tech for Facebook / Apple / Amazon / Netflix / Google).

Scope

I’m exploring other deFAANG opportunities in different parts of my digital life. The desktop computer theme is a critical one to get right. I had, for many years, trusted Apple because of their overt and unflapping sense of urgency to defend the privacy of their customers.

But how well they are doing that is becoming increasingly unclear. We now know that the United Kingdom has ordered Apple to create a back door for them (and how much or little they’ve complied or intend to comply is something we can’t audit, since the company is known for its really radical level of opacity about its internal operations). We can only assume China and other major states have asked for the same. Considering the ease with which Apple gets approvals to work in countries like China, it’s really calling into question how well they’ve lived up to their own privacy hype.

This effort is just to try to figure out if a free and open Operating System based on Linux could reasonably replace macOS in my life.

I’ll separately be evaluating other aspects of my online life, including wearables, mobile devices, IoT, media devices, etc.

Considerations

Some things I’m going to be looking for include (not an exhaustive list):

  • Does it work well with the things I do every day on a desktop computer. It’s ok if software platforms have to be replaced. In some cases, I’m really expecting that. This is mostly a lot easier than it may sound in my specific case.
  • Is it a joy to use?
  • Is it reliable and maintainable?
  • Does it show a lot of promise in terms of interoperating with the other device classes I mentioned before, assuming I shift to more open platforms there?

Setting up for success

While it would be tempting to try to run this experiment on a Raspberry Pi, I’d probably undercut my success on factors like “joy” or being able to run all of the apps I want to run (not everything is built for this architecture).

I purchased an excessively cheap mini PC, brand new, based on the Intel N100 processor. These things are well under $200 new, have more than adequate memory, fast NVMe storage. It’s not anything I can run local AI on. And not video editing, either. But the vast majority of the rest of what I do should work fine on this, and give me some indication of whether or not it’s worth investing in a “real computer” for a Linux desktop. This investment is not a waste of money, because I run a number of self-hosted services on small computers, and once I’m done with this experiment it would be easy to repurpose this computer as a “server upgrade” for one of my other mini PC’s that I’m already using. Some of them are quite old and have earned a retirement plan.

Early impressions / random thoughts

  • Fedora 41 is mostly really easy to install. Installing it over top of an existing operating sytem required more “engineer” skillset than a casual user could be expected to know. So while Fedora has come a very long way in 20+ years, there are still elements where it’s clearly not intended for casual users to take on. I have that background so I’m ok with it for my purposes.
  • I understand why the Fedora Project is very principled in its exclusion of common media codecs at install-time. But what surprises me is that it wasn’t very easy to find the right incantations to get those codecs installed and working reliably. I’m still not 100% sure that I got it all. If I move to a different distribution than Fedora, this might be the main reason why.
  • The Flatpak method of installing apps is amazing. But it’s not enabled by default. Enabling this helped me to get way more comfortable in Fedora very quickly.
  • I’m still running into weird gtk vs qt issues with some apps. It’s kinda sad that in over 25 years, it’s still weird to pull a KDE app into Gnome, or a Gnome app into KDE. Some apps won’t run at all in KDE.
  • There’s also really weird UI bugs with only some apps whereby the initial window comes up in a certain size and shape, and I can’t grab the edges with my mouse pointer to resize the windows.