I’ve planned out all the courses I’ll watch over the next 6 months.

The first course, called Docker mastery is 21 hours long. I’ve been trying to spend an hour a day on average to watch and practice.

I’m combing video lessons with docs, youtube videos, articles, perplexity and claude.

I’ve reached Docker Swarm section and I’m now learning how to orchestrate containers on multiple nodes.

I’ve also switched to Neovim and I’ll be exclusively coding on it for the whole year (except for unusual circumstances, if such will happen for unknown reasons).

I’m also re-reading Clean Architecture and tbh, I’m now understanding far more than when I partially read it the first time.

I’m refactoring Jonli Chat too, although not as fast as a month ago. But I’ll get to it.

So that’s where I’m at. And yes, I still spend more time on X and Threads than anywhere else.

This will be epic. Lex is going interview the legends

Reading a book on customers’ needs

While I'm just getting started with Docker Swarm, I know for sure that there are differences between Docker Compose and Docker Swarm.

Docker Compose essentialy automates the process of starting containers, setting up networks, building images, publishing ports and setting up environment for you, so that you don't do it every time by hand with docker container run for each container. You just write one or more yml files and declaratively describe the services you need.

Docker Compose is good for development, so that you start multiple containers that depend on each other, but it won't handle orchestrating containers on multiple nodes, load balancing them, doing blue-green deployments, replicating containers between multiple nodes, etc.

So Docker Swarm is definitely the correct choice for when you want to ensure to have multiple nodes.

For now I'm playing around with Docker Swarm locally via multipass. I do plan to try and deploy multiple DigitalOcean droplets, initialize a Swarm and try to manage managers and workers, deploy stuff and break stuff.

As for Kubernetes, what I know for now is that it's more complex and allows more stuff to be done. Docker Swarm is pretty easy to set up, it comes with sensible defaults and you don't really have to do much.

I have about 120 lessons to watch. I do have a Docker book that I want to finish as well.

The book I have is called "Docker Up & Running" and the first time I started reading the book it seemed to difficult to grasp for me. The course was definitely a much easier approach with a less steep curve.

But skimming through the book I did realize that it has a lot more interesting examples and practices.

I'm definitely not following the content goals for now. My top priorities are working on my skills

I work so much at home that I feel anxious when working outside. It came to a point where one of my goals is to sit more outside and be less socially awkward. Relax more in public, get rid of the spotlight effect, etc

One of the contributors to me working less outside is the fact the we moved further away from the city's center, where all of the cafes, coworkings are. I liked the vibe at the center more, the people, the atmosphere…here it's a bit different. It's just me with a laptop. Heck, even the coffee here is different.

Same goes with a gym, when there are a lot of people pumping, doing stuff – you get effected by that and also do stuff, take a part of their confidence and energy.

I was focused on a Docker course. Doing good progress. But I wanted to get back to working on @JonliChatBot, as the month is coming to an end.

In my dev environment I've already added profile pics and now I'm working on adding regions (top 118 countries using Telegram).

I also wanted to take a peek at what the giants are doing and what's working for them, so that I know where to shift my focus and time.

I've tried ome.tv on my phone and I have to be honest – it's one of the worst UX I've ever seen.

Nevertheless, according to public data, in December 2024 alone ome.tv's revenue was about $400 000. They use a freemium model with a ~$8 monthly subscription.

They don't have as much features as their competitors like Azar, Tinder, etc. Yes, I consider Tinder to be their indirectish competitor too.

Azar's annual revenue though (spoiler alert, also according to public data), is about $47.6 million. Yeah…

Sure, huge difference, but also goes to show that even with a shitty UX, ome.tv pulls-off a ~$4 million a year revenue.

Anyway, main features of ome.tv (based on a quck glance at their app) were connections, direct messages, filtering by gender and country, and Tinder like features.

Either they're not making $400 000 a month or they're not interested in improving their UX (for whatever reason).

Just pushed a minor update to @JonliChatBot. Let me know if you'll have issues and please send screenshots. Until then, I'll work on region and gender filtering.

I'm also considering two things:

1. Substitute the "Next chat" button with a simple swipe gesture
2. Remove the hide camera and mute microphone button (or move them somewhere else into settings)

How my setup looks right now

https://www.threads.net/@ramzcoder/post/DFDBs9MMMqD

My first popularish post in English on Threads

Even Aivars Meijers replied, who I really respect

This just motivated me even more to keep posting in English

This channel is now @ramzcoder

Have you seen DeepSeek-r1 LLM's benchmarks?

It matches ChatGPT o1 at a lower price.

Try using it, it's even FREE and can run locally via ollama.

https://deepseek.com

They also have a decent chat app

Guys, suggest a few places to work and have a cup of coffee in Tashkent