Software engineering is often viewed as a highly technical discipline characterized by coding, algorithms and data. While technical expertise is essential, there is another critical factor that is often overlooked, which is domain knowledge.
In recent years prediction markets have gained renewed attention through platforms such as Polymarket and Kalshi, which allow users to place bets on political outcomes, economic indicators and other real-world events. On the one hand, supporters argue that these markets aggregate dispersed information and produce more accurate forecasts than polls or experts. On the other hand, critics warn that prediction markets raise serious ethical concerns, especially when they involve harmful or violent events.
Within a very short period of time, and without any broader public debate, Western governments have quietly rolled out age verification laws at the operating-system level, presenting them as tools to protect children online. While the stated goal of protecting minors may sound reasonable, most people seem unaware that these proposals raise serious concerns about privacy, surveillance, population control and digital freedom. Practically speaking, they are a trojan horse for ID collection and the total digital surveillance state.
When I recently went down the OceanGate rabbithole I realized that several factors contributing to the tragedy are related to the lack of sound engineering principles, which I previously mentioned in one of my blog posts and which help me build high-quality software. The OceanGate tragedy is a fresh reminder of why sound engineering principles are essential, especially when it comes to structures whose failure can cause death and destruction.
Just recently I learned a new software tool that allows me to program and simulate a robot, using Python, ROS2 and Gazebo. This setup can be used not only to code the software for a real world robot, but also to simulate the behavior of the robot, its physics and environment.
Over the past few decades the lean production paradigm, along with the closely related just-in-time production, has dominated modern supply chain management and manufacturing systems. However, the Covid-19 pandemic revealed a fundamental weakness of modern supply chains and production systems: a persistent trade-off between efficiency and resilience.
The invention of large language models (LLMs) has profoundly shaped the landscape of AI, sparking another wave of hype in the media and the industry with voices even speculating about the emergence of general artifical intelligence. What often goes overlooked in these optimistic accounts, however, is the fact that the challenges of AI to engineer general artificial intelligence are more fundamental than commonly assumed.
In recent years software engineers have become sloppy, especially when compared to the rigor and precision of traditional engineering disciplines. And, I think, at least part of the reason is that software engineering has increasingly decoupled from physical constraints. In recent years I came to the conclusion that I could borrow principles from traditional engineering and apply them to software engineering in order to improve the quality of my software products.
The end of the year is the perfect time to reflect on past challenges and errors, draw valuable lessons and set clear goals for the future.
Just recently I tested a couple of email clients on my Ubuntu machine, one of which was Evolution - the Gnome email client. Since Evolution doesn't allow me to implement CSS/HTML code into the email, I decided to uninstall the client and its related files. When I later rebooted my system, however, I almost got a heart attack.
Engineers are often faced with the problem of trade-offs, which describes a situation, in which a certain advantage comes with a cost. In the blockchain space engineers have to deal with a fundamental trade-off between efficiency and resilience: either the blockchain is efficient and fast, which comes at the expense of its resilience or a blockchain is characterized by a high resilience, which makes it ineffcient and slow.
Just recently I developed an Android app using Python and Kivy, which required some tight security measures because it involved sensitive patient data. Moreover, the app allowed users to make requests to a server, for which a token was needed. Since I couldn't find any tutorial explaining encryption and decription for the Android OS based on Python Kivy, I decided to write a short tutorial on how to implement encryption and decryption on an Android device using Python Kivy.
Universities are considered as the most trusted institutions in Western societies – a possible reason why some readers might view this post as exaggerating, if not conspiratorial. But the sad reality is that trust in science and academic institutions is increasingly compromised. In fact, there are three main reasons why science and diplomas are not a religion and why we should embrace scepticism towards science and academia: (1) technical and scientific errors, (2) corruption and (3) a complicated relationship between science and journalism.
One of the claims stated by supporters of the Covid regime is that Covid cases are underreported and that the official statistics do not reflect reality. I agree with the latter part of the claim, but in this article I will show why the claim that Covid numbers are underreported is not true and why, in fact, the official statistics overstate the actual prevalence of C-19 to a considerable extent.
It is a common practice at universities for math and statistics professors to demonstrate models by using perfect, and in many cases even fictional data. When students later apply these models in real-world scenarios, however, they often encounter the stark reality that perfect data does not exist. Most datasets are characterized by underlying biases, many of which remain unidentified. And even when biases are recognized, there are no formulas to account for these errors.
A common problem with Javascript pages is that their html code might not contain the data that we want to crawl. Instead, the html of Javascript pages includes pieces of JS code that dynamically load data from the database in the backend into the page, but not the actual data. To deal with this problem we can use the selenium library which comes with a variety of useful functions, one of which allows us to extract data loaded from the backend.
In this post I will show how to develop a neural network by using Python and Keras. The neural network was part of an old project, in which the system extracts the statistically most relevant keywords from a website and classifies them into the respective industry category. In this post the model will be trained by a dataset with 16,000 firms and their top website keywords as input variables and more than 30 different industry categories as the output variable.
The complexity of what military pundits call cyber space consists of a mixture of a large number of technologies and tools at different levels as well as their dynamic interaction and progress. In the following blog post, I will present common offensive and defensive tools and techniques used in cyber warfare and provide a basic overview of their general principles.