The Idea My original idea was to study how to RAG (Retrieval-Augmented Generation) an LLM model by inserting something potentially useful to it. I chose qwen3-coder as my language model, and qwen3-embedding as my embedding model, both easily available via Ollama. My current work laptop is not powerful enough for any of this work, so I used my personal desktop computer for the testing, with AMD’s ROCm containers. In Practice The LLM vibes brought me to learn a couple of things.

The yearly Mindtrek 2025 was arranged again at Tampere, Finland, with some nice changes to the format. The full program was divided into three tracks – in addition to the former busines and public sector tracks, there was also a track for developers! I spend most of the time in the new Developers track thanks to interesting talks. SUSE also had a presentation about digital sovereignty, including operations sovereignty, business continuity and overall making a great case for why open source is not a plus, it’s a must.

The State at the Beginning Earlier I had made progress on my server to serve different websites from inside containers, including refreshing SSL certificates. But the server started to be aging, and for both learning and future proofing purposes I started looking at migrating to something newer. And, I never really finished all the old work anyway. Instead I have a new container in use that has SSH running inside, that should be migrated as well.

SUSE was a gold sponsor at Mindtrek 2024, a conference with a long, almost 30 years history starting as a ”multimedia competition”, always with academic conference held alongside it, and more recently held by COSS – the Finnish Centre for Open Systems and Solutions. This year Mindtrek was having a very open source focused program with two tracks, ”The Future of Open Source Business” and ”Enhancing Public Service with Open Source”.

Coincidentally I have had multiple needs for serving websites from behind containers instead of directly from the host. After some practicing, which can be also described as ”hacking without carefully documenting what I did”, I decided not to explicitly describe my setup but how I achieved (towards) my goal of serving my web sites (3 on the same server) from behind containers. This will enable me to consider using things like PHP or other notorious pieces of software thanks to a level of sandboxing.

Lately I needed to mirror a website as fully as possible, and ended up researching a bit more than my previous times I’ve done so. Here I’m just dropping a note that I ended up doing the following: wget -mkxp --adjust-extension -e robots=off https://myurl.com/ Here -m is: -r -N -l inf --no-remove-listing or in long form: --recursive --timestamping --level inf --no-remove-listing and the rest ie -kxp are, in the same order

This hackweek I’ve been playing a bit around with my desktop computer which has AMD Radeon 6600 XT graphics card which is based on the RDNA2 architecture. The idea was to find a way to utilize it for Stable Diffusion Version 2 latent text-to-image diffusion model without invading the host too much with randomly downloaded modules, but still using the GPU for computing. The graphics card has “only” 8GB RAM which is apparently only a starter amount in this field, so I needed to also check if that’s enough.

ALP - The Adaptable Linux Platform – is a new operating system from SUSE to run containerized and virtualized workloads. It is in early prototype phase, but the development is done completely openly so it’s easy to jump in to try it. For this trying out, I used the latest encrypted build – as of the writing, 22.1 – from ALP images. I imported it in virt-manager as a Generic Linux 2022 image, using UEFI instead of BIOS, added a TPM device (which I’m interested in otherwise) and referring to an Ignition JSON file in the XML config in virt-manager.

One thing I wanted to investigate during the Hackweek was trying out whether openSUSE Leap would offer my Raspberry Pi Zero 2 a nice and stable option for motion detection camera recording. I have had RPi 3 Model A+ doing this for a longer time, and a Zero model before that, but the newer RPi Zero 2 has been a bit unstable for so far unknown reason. There are also some unoptimal combinations of too old or too fresh software in the official Raspberry Pi OS releases.

I got ”inspired” by my writing of the previous blog post, and wrote in a channel about my experience some time ago. So why not also do a blog post about doing a blog post :) So… I was planning to use GitLab’s Pages feature via my Hugo fork as usual to push it through. So like, concentrate on writing and do a publish, right, like in good old times? I did so, but all I got both locally and in remote pipeline was stuff like…

I’ve always, or at least ever since the development of the iconic Nokia N9 and the projects I was working on at the time, wanted “60 fps” silky smooth behavior from both phones and computers, and learned to be sensitive to that. GNOME has been fighting back a bit on that front for several years though on HiDPI displays, with also regressing at least on openSUSE a bit earlier which I was unable to pinpoint exact reason to.

I received a new laptop for work - a Dell XPS 13. Dell has been long famous for offering certain models with pre-installed Linux as a supported option, and opting for those is nice for moving some euros/dollars from certain PC desktop OS monopoly towards Linux desktop engineering costs. Notably Lenovo also offers Ubuntu and Fedora options on many models these days (like Carbon X1 and P15 Gen 2). Obviously a smooth, ready-to-rock Ubuntu installation is nice for most people already, but I need openSUSE, so after checking everything is fine with Ubuntu, I continued to install openSUSE Tumbleweed as a dual boot option.

I’m using an external keyboard (1) and mouse (2), but the laptop lid is usually still open for better cooling. That means the internal keyboard (3) and touchpad (4) – made of comfortable materials – are open to be used by a cat searching for warmth (7), in the obvious “every time” case that a normal non-heated nest (6) is not enough. The problem is, everything goes chaotic at that point in the default configuration.

This is more like a self-written notes post about a problem I’m facing, since my laptop’s web camera is starting to deteriorate. I’ll update the post if I find more useful bits of how to tweak with USB. For the first few weeks I was suspecting a Tumbleweed problem, but eventually I booted up Ubuntu LTS from USB stick and managed to see the problem also there. dmesg goes like follows

Google Pixel phones support what they call ”Motion Photo” which is essentially a photo with a short video clip attached to it. They are quite nice since they bring the moment alive, especially as the capturing of the video starts a small moment before the shutter button is pressed. For most viewing programs they simply show as static JPEG photos, but there is more to the files. I’d really love proper Shotwell support for these file formats, so I posted a longish explanation with many of the details in this blog post to a ticket there too.