Just wrapped up the first day of of J On The Beach , a conference that takes place in my hometown (8 years and counting) but I had never attended before.
The main question coming into day 1 was to test how bad an idea was to take 3
different workshops on 3 different topics which I have little experience with.
I did survive, and finished the day with more energy that I started, so I
guess curb boost your enthusiasm.
These are the 3 I attended:
I don't do any fancy things with Docker, but I love low-level unixy stuff, so I thought I would check this one out. I did learn quite a few things. My favorite ones were:
docker history it will show you everything that has happened in the build
process (and why it's a good idea to include a step to build a new image out of
your finished image, so you minimize the risk of exposing sensitive information)Here are some materials I used to prep for the workshop. These formed the basis of the material covered
As much as decentralization and p2p has been co-opted by some sector of tech (i won't mention any key term for fear of getting more spambots), it's still an interesting and valuable technology. I loved seeing how Rüdiger claimed that type of pragmatism explictly when motivating the workshop.
He spent a good portion of the workshop explaining their technical philosophy, and the underlying networking concepts. I went home with a bunch of networking items to explore further (QUIC, Nat hole punching, ALPN, ...).
Given that most of the participants (me included) were not very familiar with rust (In my case I had also just broke my vim rust setup by installing Coc Rust Analyzer 😢), the hands on part was more like a walkthrough of the code, but the incremental demos were super helpful to understanding the concepts.
I've been using rust in the last few months and as hard as it sometimes is, this seems like a great opportunity to continue with it.
This was the more enterprisey / vendor-centric one, but I have been curious about AI Agents, Langchain and Vector DBs for quite a while. Luckily it started with a good half hour of theory. I still don't fully get how meaning is captured/derived from embeddings, but I know a little bit more.
The workshop leveraged managed services for everything, which intiially bummed me but it's amazing how much you can accomplish without touching a line of code. I particularly loved Streamlit, which I had never heard of before.
I checked this tutorial just before the workshop, and it helped with context.
It's been quite interesting to experience the different types of workshops, but this is going to be it for now. Let's see what the full conference has to offer.