In the vast landscape of business, one truth remains immutable: you can’t negotiate genuine desire. It’s a lesson that traditional industries have tried to bend to their will, and even in the dynamic world of startups and tech products, it’s a concept that refuses to yield.
In the realm of software development, crafting code that remains maintainable, flexible, and scalable stands as an utmost priority. And when delving into the concept of scalability, I’m considering not only the system’s load capacity but also your ability to maintain it, essentially wielding it as a tool to conquer complexity. To achieve this, developers often embrace design principles that steer them in writing this type of high-quality code. One such set of principles is known as SOLID, an acronym representing five essential tenets in object-oriented programming. ...
In my previous article, How to Create a Pharo Smalltalk Plugin, I outlined how to create a Pharo Smalltalk plugin. However, since then, I’ve discovered a more convenient way to produce these plugins using the PharoPluginBuilder and StarterPlugin. In this post, I wanted to inform and describe the idea of this simplified process of creating a Pharo Smalltalk plugin.
As businesses increasingly rely on software to deliver their products and services, ensuring that their production applications are running smoothly and efficiently becomes a must for maintaining what makes your app to be dependable on: application reliability, performance, and availability. One critical aspect of monitoring production apps is measuring resource usage, including CPU, memory, and network utilization and that’s very well covered by many solutions. As previously mentioned, I particularly like The TLIG Stack for that. These metrics can provide valuable insights into the application’s health, allowing teams to identify performance bottlenecks and detect potential issues before they escalate into larger problems. ...
I’m excited to announce that the Pharo Consortium has been approved as an organizer for Google Summer of Code 2023! As a mentor for the Pharo TensorFlow plugin project, I’m thrilled to be a part of this program and to help developers to work on a project for the Pharo community.
Monitoring is an essential part of any modern infrastructure. In this article, I will show you how to set up Promtail, Telegraf, InfluxDB, Loki, and Grafana to monitor your system’s performance and logs and use Kuma for availability and alerts. I’ve selected this toolchain because is reasonably straightforward to setup, and shows good potential and flexibility for further growing needs.
Pharo is a powerful open-source Smalltalk-based programming language and environment that has been widely used for research and development in various fields. Specially Pharo at Inria, France’s National Institute for Research in Digital Science and Technology. However, the lack of a native plugin for TensorFlow, a popular open-source machine learning framework, has been a major limitation for high efficiency AI development with Pharo. I’ve written this post to let you know that, I’ve proposed a Google Summer of Code 2023 project to create a TensorFlow plugin for Pharo, which will unlock the full potential of Pharo for production grade AI development. ...
When working with Smalltalk, you may wish to access functionality that exists in a specific library or technology that you do not want to recreate. In such cases, you have two options: creating an FFI bridge or extending the Smalltalk VM with new primitives. In this article, we’ll explore the less common method of extending the Pharo Virtual Machine with an external plugin by creating a HelloWorldPlugin. The goal is to provide insight and clarity into the process of extending the VM in this manner, and to explain the steps involved in Pharo 9 as of February 2023.
Today, I’m excited to announce a new release of PharoDawnTheme, this dark warm color theme for the Pharo Smalltalk IDE. This theme is available for Pharo 9 and 10
Just a quick update to mention that I’ve merged in develop a pull request that will add the capability to work with Mapless using UnQLite in memory. The use case is mostly for single-image caching. When you want to cache data but keep the image lean, that’s when you can use this setup.