In my post last month about rethinking my personal tech stack, I had mentioned that I had been using Ente Photos as an alternative to Google Photos, but had also been evaluating Immich. I also mentioned that one day in the future I might move over to Immich after figuring out a better backup strategy for my HomeLab.

That day came this month as I had some time to figure out the last few items I needed to feel secure in using Immich as my primary photo platform, running entirely on my Beelink S12 Pro running Proxmox.

Several years ago, I bought a Synology DS920+ to use as a network attached storage (NAS) appliance in the house. I don't work as a photographer or video editor so my work doesn't really have any massive storage needs; rather, I wanted someplace to host media for the family that didn't require an ongoing subscription service to access. For a few years, it served that purpose almost exclusively, both as a drive I could mount and as a Plex server.

Recently, I had begun tinkering with a few other services and seeing how far I could push my Synology. It has a Container Manager service that allows you to run Docker containers, although the interface is very GUI-driven and there's always a few gotchas. For example, Synology reserves port 80 and 443 for their own web station and reverse proxy services, which makes it difficult to run your own reverse proxy on the device.

For these reasons, I started to consider buying a separate device to act as a home server and repository for all my containers that I would have full control over, and relegate the Synology back to its role as just dumb storage.

Like many people, I usually take the start of a new year as an opportunity to reevaluate some choices and habits I have and see if there is room for improvement. One such area up for review is the set of services, apps, and subscriptions I use outside of work.

Much of my current computing habits can trace their origins back to the 2000’s, where Gmail marked the start of Google’s growth beyond just search and ads and smartphones began to take off. That was a very long time ago, and a lot has changed in both my own habits and the tech industry since then - so I used most of 2025 to begin researching and evaluating alternatives to what I was currently using. I placed a point on emphasis on identifying options that were cross-platform (avoiding lock in to any one operating system) and gave preference to alternatives that were open source and/or allowed me to either self host or at least keep a copy of my data synced locally, in case of account lockout. I ended up finding a lot of great options, some of which I’ve switched over to entirely.

This post will highlight some of these alternatives - I plan on making more in depth posts about each one specifically at some point in the future.

Recently, Ruby Central, the non-profit responsible for maintaining the RubyGems package manager, suffered a security incident where they temporarily lost control of their AWS account. They have since published a post-mortem of this event that is ostensibly aimed at putting their community's minds at ease. Unfortunately, it had the opposite effect on me, causing me to come away with more questions than answers.

While AWS re:Invent doesn't officially kick off until December 2, we are now officially in the lead up to the event, a time that often seea a flurry of new feature and service announcements. While these announcements are often overshadowed by new products that debut at the conference, they are often just as important (if not more so) to the work I do every day. This week has been the strongest example of this trend to date, with several announcements that made me genuinely excited. In the interest of processing that enthusiasm, I decided to write a recap of what I view to be the biggest ones.

Recently, I had to complete a project that involved running the open source tool Cloudquery to create an inventory of resources in a GCP organization. This assignment turned out to be a great introduction to learning how Google Cloud Platform works, as I had almost exclusively used AWS previously (with only minor trials in both Azure and GCP during that time). Throughout the project, I found myself often mapping certain concepts back to what they would be called or how they would be done in AWS. In doing so, I started to keep score of what I liked more than AWS and what I liked less - and now I'm finally sitting down to organize some of those thoughts.

The annual AWS re:Invent conference has come and gone. As usual, there is an overwhelming amount of new product launches, feature enhancements, and other service offering announcements to parse through. You could spend several days just sifting through all of the information on 'new stuff'. What I was interested this time, however, was some of the announcements for security-related services and features, especially those that solve pain points I've experienced in the past.