Today you can read about, among others, the effectiveness of doing something every day, Java virtual threads (coming soon!), best practices for deploying Java apps on Kubernetes, finding a bug in the Google Pixel Markup tool and visual story from NYTimes about how ChatGPT could put ‘watermark’ in generated text!

Business and (side)project section

1. The unreasonable effectiveness of just showing up everyday

You may tell “I do not have time to create a new feature for my product”, but what if you will set yourself a simple rule - “do something every day”? The author of Typesense (open source search) does that and it proves that he delivered a lot during six years of small additions every day!

2. My experience releasing 3 failed SaaS products

How it is to create an app, release it and face a situation when no-one is using it? Some of the lessons learnt include “MVP doesn’t need to be pretty, but must solve customer problem” or “coding is the easiest thing in SaaS”. Worth reading!

Developer section

3. Best Practices for Java Apps on Kubernetes

One of the common deployment ways of applications is deploying them on Kubernetes. In the linked article you can find best practices to deploy Java applications (including, but not limited to, Spring Boot and Quarkus based ones) on the K8s cluster.

4. The Ultimate Guide to Java Virtual Threads

Virtual Threads are coming to Java! They are still in the preview phase, but you can check them out already with JDK19 and enable them manually. The final version of the features might differ from the current state, but it is good for Java devs to see what is coming!

5. My Setup for Self-Hosting Dozens of Web Applications + Services on a Single Server

What is the simplest solution to host multiple applications? Just do it on a single server! The author presents how configured his (rented from OVH) single server to host all applications that he created.

6. Exploiting aCropalypse: Recovering Truncated PNGs

Have you ever used the Markup tool (app for cropping and drawing over the screenshot) on Google Pixel? If the answer is yes then you are affected - your actions on screenshots can be in a lot of cases reverted. The blog post tells a story of finding the bug and explains the possible cause (possible as this part of the code is not open-sourced).

7. How ChatGPT Could Embed a ‘Watermark’ in the Text It Generates

Recognition of the text generated by LLMs (Large Language Models) starts to be a serious task - especially in education. New York Times reporter prepares an interactive example of how ChatGPT could embed a watermark in the generated text to make this possible. Also nice, in a story way, explanation of how ChatGPT works.