All streams

We continue the series of articles about new patches coming to PostgreSQL 18, this one covering the news of the November CommitFest.

If you missed the previous reviews, you can check them out here: 2024-07, 2024-09.

Since sharing my initial thoughts on Google's Gemini 2.5 Pro, I've dug even deeper and gathered answers to many common questions developers like us might have. This AI model is making waves, so let's break down everything you need to know – from its groundbreaking features to the practical stuff like cost and limitations.

This world needs a new theory — a theory that could describe all the theories on the planet. A theory that could easily describe philosophy, mathematics, physics, and psychology. The one that makes all kinds of sciences computable.

This is exactly what we are working on. If we succeed, this theory will become the unified meta-theory of everything.

A year has passed since our last publication, and our task is to share the progress with our English-speaking audience. This is still not a stable version; it’s a draft. Therefore, we welcome any feedback, as well as your participation in the development of the links theory.

As with everything we have done before, the links theory is published and released into the public domain — it belongs to humanity, that means, it is yours. This work has many authors, but the work itself is far more important than any specific authorship. We hope that today it can become useful to more people.

We invite you to become a part of this exciting adventure.

Google has introduced Gemma 3, a collection of lightweight, open-source AI models designed for efficient performance across various devices, from smartphones to workstations. Building upon the technology of the Gemini 2.0 models, Gemma 3 supports over 140 languages and offers advanced text and visual reasoning capabilities. What's to know more? Then, read this article until the end.

Automatic data scraping (parsing) has become an essential practice for developers, analysts, and automation specialists. It is used to extract massive amounts of information from websites—from competitors’ prices and reviews to social media content. To achieve this, numerous “scrapers” have been developed—libraries, frameworks, and cloud services that enable programmatic extraction of web data. Some solutions are designed for rapid parsing of static pages, others for bypassing complex JavaScript navigation, and yet others for retrieving data via APIs.

In this article, I will review the top scraping tools—both open source libraries and commercial SaaS/API services—and compare them according to key metrics: • Speed and scalability; • Ability to bypass anti-bot protections; • Proxy support and CAPTCHA recognition; • Quality of documentation; • Availability of APIs and other important features.

The world of AI is constantly evolving, and it feels like every week brings a new tool promising to revolutionize how we work and create. I recently got a chance to try out Gemini Canvas, a new interactive space built into Google’s Gemini app. If you want to know what I discovered and how you might use it, read this article until the end.

The most common types of software bugs are memory management bugs. And very often they lead to the most tragic consequences. There are many types of memory bugs, but the only ones that matter now are memory leaks due to circular references, when two or more objects directly or indirectly refer to each other, causing the RAM available to the application to gradually decrease because it cannot be freed.

Memory leaks due to circular references are the most difficult to analyze, while all other types have been successfully solved for a long time. All other memory bugs can be solved at the programming language level (for example, with garbage collectors, borrow checking or library templates), but the problem of memory leaks due to circular references remains unsolved to this day.

But it seems to me that there is a very simple way to solve the problem of memory leaks due to circular references in a program, which can be implemented in almost any typed programming language, of course, if you do not use the all-permissive keyword unsafe for Rust or std::reinterpret_cast in the case of C++.

If you’re building a multilingual React Native (or web) app, you’ve probably tried react-i18next, i18n-js, LinguiJS, or similar libraries.

But in every project, the same issues come up:

❌ Unused key-value pairs are never removed
❌ Content gets duplicated
❌ Ensuring format consistency across languages is painful
❌ i18next doesn’t generate TypeScript types by default – so t("my.key") won’t throw even if it’s been deleted
❌ Localization platforms like Lokalise or Locize get expensive fast

Frustrated by these challenges, I waited for a better solution... then decided to build one myself: Intlayer.

If your code has many nested executions of stored procedures, you can benefit from building popular "flame diagram" of the execution time which is de facto standard for performance profiling.

Storing all your data in one place might seem convenient, but it’s often impractical. High costs, database scalability limits, and complex administration create major hurdles. That’s why smart businesses rely on Information Lifecycle Management (ILM) — a structured approach that automates data management based on policies and best practices.

With Postgres Pro Enterprise 17, ILM is now easier than ever, thanks to the pgpro_ilm extension. This tool enables seamless data tiering, much like Oracle's ILM functionality. Let’s dive into the challenges of managing large databases, how ILM solves them, and how you can implement it in Postgres Pro Enterprise 17.

Check Yourself:

– Does your mindset hold you back?
– Do you try to look smarter than you are?
– Do you avoid difficult tasks to not seem "incompetent"?
– Do you believe success is all about talent, not effort?
– Do you think working harder means you're less talented?

If you said yes to these,

Hey everyone! I'm super excited to share something cool I've been playing around with: Google AI Studio. It's like a playground where you can build stuff with Google's AI models. So, if you're curious and want to dip your toes into the world of AI, follow along! I'll show you the basics.

Author: Sergey Lukyanchikov, C-NLTX/Open-Source

Disclaimer: The views expressed in this document reflect the author's subjective perspective on the current and potential capabilities of jBPM.

This text presents jBPM as a platform for orchestrating external AI-centric environments, such as Python, used for designing and running AI solutions. We will provide an overview of jBPM’s most relevant functionalities for AI orchestration and walk you through a practical example that demonstrates its effectiveness as an AI orchestration platform:

How did engineers in the past manage to measure electrical power without modern microchips and DSPs? This article explores the Energomera CE6806P, a device created in 2006 for verifying electricity meters, yet built using 1980s-era technology.

We’ll take a closer look at its design, principles of operation, and how discrete-analog solutions were used to achieve high accuracy. The Energomera is a fascinating example of engineering and ingenuity, giving us a unique perspective on the evolution of electrical measurement devices.

While pg_probackup 3 is still in the works and not yet available to the public, let’s dive into what’s new under the hood. There’s a lot to unpack — from a completely reimagined application architecture to long-awaited features and seamless integration with other tools. 

A lot of people around me spend time trading on the stock market. Some trade crypto, some trade stocks, others trade currencies. Some call themselves investors, others call themselves traders. I often see random passersby in various cities and countries checking their trading terminals on their phones or laptops. And at night I sometimes write analytical or backtesting software—well, I did up until recently. All these people share a common faith and a set of misconceptions about the market.

In this article, I'll take you through everything you need to know about Hugging Face—what it is, how to use it, and why it's a game-changer in the ever-evolving landscape of artificial intelligence. Whether you're a seasoned data scientist or an enthusiastic beginner eager to dive into AI, the insights shared here will equip you with the knowledge to Hugging Face's full potential.

Lately, tons of new Telegram channels, bots, and mini-apps in English have been popping up. Just a year ago, this trend was barely starting — most English-speaking users couldn’t even tell the difference between a Telegram channel and a group.

And now? They’re all in, growing their channels, bots, and mini-apps like crazy. Telegram is turning into a massive platform.

Often at work, I encounter services that provide offerings such as resident proxies. Yet, I have never delved deeply into the topic. I have always simply consumed the product “as is,” as some lazy authors like to say.

I have a general understanding of how this type of service works at a layman’s level, and I became interested in exploring the topic more deeply and attempting to share the conclusions I reached through a deeper understanding of what resident proxies are. Let’s see what comes out of it. No recommendations here—just the subjective, evaluative opinion of yet another “specialist.”

Proxy servers are intermediaries between your device and the internet, allowing you to hide your real IP address and alter the appearance of your connection. Think of it as a white camouflage coat in snowy weather, if we speak in very simplistic terms. Let’s start from that—options for camouflage. However, comparing with camouflage coats would be rather dull; instead, let’s recall animals and insects that use camouflage and try to draw a parallel. In fact, I’ve already done so.

For many years, the PostgreSQL community was skeptical about using this database management system (DBMS) for high-transaction environments. While PostgreSQL worked well for lab tests, mid-tier web applications, and smaller backend systems, it was believed that for heavy transactional loads, you’d need an expensive DBMS designed specifically for such purposes. As a result, PostgreSQL wasn’t particularly developed in that direction, leaving a range of issues unanswered.

However, the reality has turned out differently. More and more of our clients are encountering problems that stem from this mindset. For example, in the global PostgreSQL community, it’s considered that 64 cores is the maximum size of a server where PostgreSQL can run effectively. But we’re now seeing that this is becoming a minimum typical configuration. One particular bottleneck that has emerged is the transaction counter, and this is a far more interesting issue. So, let’s dive into what the problem is, how we solved it, and what the international community thinks about it.